OP5-Log-Analytics-6.x User Guide

About

OP5 Log Analytics User Guide

Software ver. 6.x

Document version. 1.0

Introduction

OP5 Log Analytics is innovation solution allowing for centralize IT systems events. It allows for an immediately review, analyze and reporting of system logs - the amount of data does not matter. OP5 Log Analytics is a response to the huge demand for storage and analysis of the large amounts of data from IT systems. OP5 Log Analytics is innovation solution that responds to the need of effectively processing large amounts of data coming from IT environments of today’s organizations. Based on the open-source project Elasticsearch valued on the marked, we have created an efficient solution with powerful data storage and searching capabilities. The System has been enriched of functionality that ensures the security of stored information, verification of users, data correlation and visualization, alerting and reporting.

OP5 Log Analytics project was created to centralize events of all IT areas in the organization. We focused on creating a tool that functionality is most expected by IT departments. Because an effective licensing model has been applied, the solution can be implemented in the scope expected by the customer even with very large volume of data. At the same time, the innovation architecture allows for servicing a large portion of data, which cannot be dedicated to solution with limited scalability.

Elasticsearch

Elasticsearch is a NoSQL database solution that is the heart of our system. Text information send to the system, application and system logs are processed by Logstash filters and directed to Elasticsearch. This storage environment creates, based on the received data, their respective layout in a binary form, called a data index. The Index is kept on Elasticsearch nodes, implementing the appropriate assumptions from the configuration, such as:

• Replication index between nodes,
• Distribution index between nodes.

The Elasticsearch environment consists of nodes:

• Data node - responsible for storing documents in indexes,
• Master node - responsible for the supervisions of nodes,
• Client node - responsible for cooperation with the client.

Data, Master and Client elements are found even in the smallest Elasticsearch installations, therefore often the environment is referred to as a cluster, regardless of the number of nodes configured. Within the cluster, Elasticsearch decides which data portions are held on a specific node.

Index layout, their name, set of fields is arbitrary and depends on the form of system usage. It is common practice to put data of a similar nature to the same type of index that has a permanent first part of the name. The second part of the name often remains the date the index was created, which in practice means that the new index is created every day. This practice, however, is conventional and every index can have its own rotation convention, name convention, construction scheme and its own set of other features. As a result of passing document through the Logstash engine, each entry receive a data field, which allow to work witch data in relations to time.

The Indexes are built with elementary part called shards. It is good practice to create Indexes with the number of shards that is the multiple of the Elasticsearch data nodes number. Elasticsearch in 6.x version has a new feature called Sequence IDs that guarantee more successful and efficient shard recovery.

Elasticsearch use the mapping to describes the fields or properties that documents of that type may have. Elasticsearch in 6.x version restrict indices to a single type.# Kibana #

Kibana

Kibana lets you visualize your Elasticsearch data and navigate the Elastic Stack. Kibana gives you the freedom to select the way you give shape to your data. And you don’t always have to know what you’re looking for. Kibana core ships with the classics: histograms, line graphs, pie charts, sunbursts, and more. Plus, you can use Vega grammar to design your own visualizations. All leverage the full aggregation capabilities of Elasticsearch. Perform advanced time series analysis on your Elasticsearch data with our curated time series UIs. Describe queries, transformations, and visualizations with powerful, easy-to-learn expressions. Kibana 6.x has two new feature - a new “Full-screen” mode to viewing dashboards, and new the “Dashboard-only” mode which enables administrators to share dashboards safely.# Logstash #

Logstash

Logstash is an open source data collection engine with real-time pipelining capabilities. Logstash can dynamically unify data from disparate sources and normalize the data into destinations of your choice. Cleanse and democratize all your data for diverse advanced downstream analytics and visualization use cases.

While Logstash originally drove innovation in log collection, its capabilities extend well beyond that use case. Any type of event can be enriched and transformed with a broad array of input, filter, and output plugins, with many native codecs further simplifying the ingestion process. Logstash accelerates your insights by harnessing a greater volume and variety of data.

Logstash 6.x version supports native support for multiple pipelines. These pipelines are defined in a pipelines.yml file which is loaded by default. Users will be able to manage multiple pipelines within Kibana. This solution uses Elasticsearch to store pipeline configurations and allows for on-the-fly reconfiguration of Logstash pipelines.# ELK #

ELK

“ELK” is the acronym for three open source projects: Elasticsearch, Logstash, and Kibana. Elasticsearch is a search and analytics engine. Logstash is a server‑side data processing pipeline that ingests data from multiple sources simultaneously, transforms it, and then sends it to a “stash” like Elasticsearch. Kibana lets users visualize data with charts and graphs in Elasticsearch. The Elastic Stack is the next evolution of the ELK Stack.

Data source and application management

Data source

Where does the data come from?

OP5 Log Analytics is a solution allowing effective data processing from the IT environment that exists in the organization.

The Elsasticsearch engine allows building a database in witch large amounts of data are stored in ordered indexes. The Logstash module is responsible for load data into Indexes, whose function is to collect data on specific tcp/udp ports, filter them, normalize them and place them in the appropriate index. Additional plugins, that we can use in Logstash reinforce the work of the module, increase its efficiency, enabling the module to quick interpret data and parse it.

Below is an example of several of the many available Logstash plugins:

exec - receive output of the shell function as an event;

imap - read email from IMAP servers;

jdbc - create events based on JDC data;

jms - create events from Jms broker;

Both Elasticsearch and Logstash are free Open-Source solutions.

More information about Elasticsearch module can be find at: https://github.com/elastic/elasticsearch

List of available Logstash plugins: https://github.com/elastic/logstash-docs/tree/master/docs/plugins

System services

For proper operation OP5 Log Analytics requires starting the following system services:

• elasticsearch.service - we can run it with a command:

    systemctl start elasticsearch.service
  

  we can check its status with a command:

  systemctl status elasticsearch.service

• kibana.service - we can run it with a command:

    systemctl start kibana.service
  

  we can check its status with a command:

  systemctl status kibana.service

• logstash.service - we can run it with a command:

    systemctl start logstash.service
  

  we can check its status with a command:

  systemctl status logstash.service

First configuration steps

Run the instalation

To install and configure OP5 Log Analytics on the CentOS Linux system you should:

• copy archive OP5 Log Analytics tar.bz2 to the hosted server;

• extract archive OP5 Log Analytics tar.bz2 contain application:

  cd /root/ tar xvfj archive.tar.bz2

• go to the application directory and run installation script as a root user:

    cd /root/insatll
    ./install.sh
  

Installation steps

During installation you will be ask about following tasks:

• add firewall exception on ports 22(ssh), 5044, 5514 (Logstash), 5601 (Kibana), 9200 (Elastisearch), 9300 (ES cross-JVM);
• installation of Java environment (Open-JDK), if you use your own Java environment - answer “N”;
• installation of Logstash application;
• configuration of Logstash with custom OP5 Log Analytics configuration;
• connect to the OP5 CentOS repository, which provides Python libraries, and some fonts;
• installation of Kibana, the OP5 Log Analytics GUI;
• installation of Python dependencies;
• installation of mail components for OP5 Log Analytics notification;
• installation of data-node of Elasticsearch;
• configuration of Elasticsearch as Data Node;
• configuration of Elasticsearch as Master Node.

Optional installation steps:

Optionally you can:

• install and configure the filebeat agent;

• install and configure the winlogbeat agent;

• configure op5 perf_data to integrated with the OP5 Monitor;

• configure naemonLogs to integrated with the Naemon;

• configure integration with Active Directory and SSO servers. You can find necessary information in 12-00-00-Integration_with_AD and 13-00-00-Windows-SSO;

• install and configure monitoring with Marvel:

    cd /usr/share/elasticsearch
    sudo bin/plugin install license
    sudo bin/plugin install marvel-agent
    systemctl restart elasticsearch
  

• enable predictive functionality in Intelligence module:

    curl -XPOST 'http://localhost:9200/_aliases' -d '{
     		"actions" : [
         	{ "add" : { "index" : "intelligence", "alias" : "predictive" } },
        	 { "add" : { "index" : "perfdata-linux", "alias" : "predictive" } }
     		]}'
  

• generate writeback index for Alert service:

    */opt/alert/bin/elastalert-create-index --config /opt/alert/config.yaml*
  

First login

If you log in to OP5 Log Analytics for the first time, you must specify the Index to be searched. We have the option of entering the name of your index, indicate a specific index from a given day, or using the asterix (*) to indicate all of them matching a specific index pattern. Therefore, to start working with OP5 Log Analytics application, we log in to it (by default the user: logserver/password:logserver).

After logging in to the application click the button “Set up index pattern” to add new index patter in Kibana:

In the “Index pattern” field enter the name of the index or index pattern (after confirming that the index or sets of indexes exists) and click “Next step” button.

In the next step, from drop down menu select the “Time filter field name”, after witch individual event (events) should be sorter. By default the timestamp is set, which is the time of occurrence of the event, but depending of the preferences. It may also be the time of the indexing or other selected based on the fields indicate on the event.

At any time, you can add more indexes or index patters by going to the main tab select „Management” and next select „Index Patterns”.

Index selection

After login into OP5 Log Analytics you will going to „Discover” tab, where you can interactively explore your data. You have access to every document in every index that matches the selected index patterns.

If you want to change selected index, drop down menu with the name of the current object in the left panel. Clicking on the object from the expanded list of previously create index patterns, will change the searched index.

Changing default users for services

Change Kibana User

Edit file /etc/systemd/system/kibana.service

	User=newuser
	Group= newuser

Edit /etc/default/kibana

	user=" newuser "
	group=" newuser "

Add appropriate permission:

	chown newuser: /usr/share/kibana/ /etc/kibana/ -R

Change Elasticsearch User

Edit /usr/lib/tmpfiles.d/elasticsearch.conf and change user name and group:

	d /var/run/elasticsearch 0755 newuser newuser – 

Create directory:

	mkdir /etc/systemd/system/elasticsearch.service.d/

Edit /etc/systemd/system/elasticsearch.service.d/01-user.conf

	[Service]
	User=newuser
	Group= newuser

Add appropriate permission:

	chown -R newuser: /var/lib/elasticsearch /usr/share/elasticsearch /etc/elasticsearch /var/log/elasticsearch

Change Logstash User

Create directory:

	mkdir /etc/systemd/system/logstash.service.d

Edit /etc/systemd/system/logstash.service.d/01-user.conf

	[Service]
	User=newuser
	Group=newuser

Add appropriate permission:

	chown -R newuser: /etc/logstash /var/log/logstash

Custom installation the OP5 Log Analytics

If you need to install OP5 Log Analytics in non-default location, use the following steps.

1. Define the variable INSTALL_PATH if you do not want default paths like “/”

    export INSTALL_PATH="/"
   

2. Install the firewalld service

    yum install firewalld
   

3. Configure the firewalld service

    systemctl enable firewalld
    systemctl start firewalld
    firewall-cmd --zone=public --add-port=22/tcp --permanent
    firewall-cmd --zone=public --add-port=443/tcp --permanent
    firewall-cmd --zone=public --add-port=5601/tcp --permanent
    firewall-cmd --zone=public --add-port=9200/tcp --permanent
    firewall-cmd --zone=public --add-port=9300/tcp --permanent
    firewall-cmd --reload
   

4. Install and enable the epel repository

    yum install epel-release
   

5. Install the Java OpenJDK

    yum install java-1.8.0-openjdk-headless.x86_64
   

6. Install the reports dependencies, e.g. for mail and fonts

    yum install fontconfig freetype freetype-devel fontconfig-devel libstdc++ urw-fonts net-tools ImageMagick ghostscript poppler-utils
   

7. Create the nessesery users acounts

    useradd -M -d ${INSTALL_PATH}/usr/share/kibana -s /sbin/nologin kibana
    useradd -M -d ${INSTALL_PATH}/usr/share/elasticsearch -s /sbin/nologin elasticsearch
    useradd -M -d ${INSTALL_PATH}/opt/alert -s /sbin/nologin alert
   

8. Remove .gitkeep files from source directory

    find . -name ".gitkeep" -delete
   

9. Install the Elasticsearch 6.2.4 files

    /bin/cp -rf elasticsearch/elasticsearch-6.2.4/* ${INSTALL_PATH}/
   

10. Install the Kibana 6.2.4 files

     /bin/cp -rf kibana/kibana-6.2.4/* ${INSTALL_PATH}/
    

11. Configure the Elasticsearch system dependencies

     /bin/cp -rf system/limits.d/30-elasticsearch.conf /etc/security/limits.d/
     /bin/cp -rf system/sysctl.d/90-elasticsearch.conf /etc/sysctl.d/
     /bin/cp -rf system/sysconfig/elasticsearch /etc/sysconfig/
     /bin/cp -rf system/rsyslog.d/intelligence.conf /etc/rsyslog.d/
     echo -e "RateLimitInterval=0\nRateLimitBurst=0" >> /etc/systemd/journald.conf
     systemctl daemon-reload
     systemctl restart rsyslog.service
     sysctl -p /etc/sysctl.d/90-elasticsearch.conf
    

12. Configure the SSL Encryption for the Kibana

     mkdir -p ${INSTALL_PATH}/etc/kibana/ssl
     openssl req -x509 -nodes -days 365 -newkey rsa:2048 -sha256 -subj '/CN=LOGSERVER/subjectAltName=LOGSERVER/' -keyout ${INSTALL_PATH}/etc/kibana/ssl/kibana.key -out ${INSTALL_PATH}/etc/kibana/ssl/kibana.crt
    

13. Install the Elasticsearch-auth plugin

     cp -rf elasticsearch/elasticsearch-auth ${INSTALL_PATH}/usr/share/elasticsearch/plugins/elasticsearch-auth
    

14. Install the Elasticsearh configuration files

     /bin/cp -rf elasticsearch/*.yml ${INSTALL_PATH}/etc/elasticsearch/
    

15. Install the Elasticsicsearch system indices

     mkdir -p ${INSTALL_PATH}/var/lib/elasticsearch
     /bin/cp -rf elasticsearch/nodes ${INSTALL_PATH}/var/lib/elasticsearch/
    

16. Add necessary permission for the Elasticsearch directories

    chown -R elasticsearch:elasticsearch ${INSTALL_PATH}/usr/share/elasticsearch ${INSTALL_PATH}/etc/elasticsearch ${INSTALL_PATH}/var/lib/elasticsearch ${INSTALL_PATH}/var/log/elasticsearch
    

17. Install the Kibana plugins

     cp -rf kibana/plugins/* ${INSTALL_PATH}/usr/share/kibana/plugins/
    

18. Extrace the node_modules for plugins and remove archive

     tar -xf ${INSTALL_PATH}/usr/share/kibana/plugins/node_modules.tar -C ${INSTALL_PATH}/usr/share/kibana/plugins/
     /bin/rm -rf ${INSTALL_PATH}/usr/share/kibana/plugins/node_modules.tar
    

19. Install the Kibana reports binaries

     cp -rf kibana/export_plugin/* ${INSTALL_PATH}/usr/share/kibana/bin/
    

20. Create directory for the Kibana reports

     /bin/cp -rf kibana/optimize ${INSTALL_PATH}/usr/share/kibana/
    

21. Install the python dependencies for reports

     tar -xf kibana/python.tar -C /usr/lib/python2.7/site-packages/
    

22. Install the Kibana custom sources

     /bin/cp -rf kibana/src/* ${INSTALL_PATH}/usr/share/kibana/src/
    

23. Install the Kibana configuration

     /bin/cp -rf kibana/kibana.yml ${INSTALL_PATH}/etc/kibana/kibana.yml
    

24. Generate the iron secret salt for Kibana

     echo "server.ironsecret: \"$(</dev/urandom tr -dc _A-Z-a-z-0-9 | head -c32)\"" >> ${INSTALL_PATH}/etc/kibana/kibana.yml
    

25. Remove old cache files

     rm -rf ${INSTALL_PATH}/usr/share/kibana/optimize/bundles/*
    

26. Install the Alert plugin

     mkdir -p ${INSTALL_PATH}/opt
     /bin/cp -rf alert ${INSTALL_PATH}/opt/alert
    

27. Install the AI plugin

     /bin/cp -rf ai ${INSTALL_PATH}/opt/ai
    

28. Set the proper permissions

     chown -R elasticsearch:elasticsearch ${INSTALL_PATH}/usr/share/elasticsearch/
     chown -R alert:alert ${INSTALL_PATH}/opt/alert
     chown -R kibana:kibana ${INSTALL_PATH}/usr/share/kibana ${INSTALL_PATH}/opt/ai ${INSTALL_PATH}/opt/alert/rules ${INSTALL_PATH}/var/lib/kibana
     chmod -R 755 ${INSTALL_PATH}/opt/ai
     chmod -R 755 ${INSTALL_PATH}/opt/alert
    

29. Install service files for the Alert, Kibana and the Elasticsearch

     /bin/cp -rf system/alert.service /usr/lib/systemd/system/alert.service
     /bin/cp -rf kibana/kibana-6.2.4/etc/systemd/system/kibana.service /usr/lib/systemd/system/kibana.service
     /bin/cp -rf elasticsearch/elasticsearch-6.2.4/usr/lib/systemd/system/elasticsearch.service /usr/lib/systemd/system/elasticsearch.service
    

30. Set property paths in service files ${INSTALL_PATH}

     perl -pi -e 's#/opt#'${INSTALL_PATH}'/opt#g' /usr/lib/systemd/system/alert.service
     perl -pi -e 's#/etc#'${INSTALL_PATH}'/etc#g' /usr/lib/systemd/system/kibana.service
     perl -pi -e 's#/usr#'${INSTALL_PATH}'/usr#g' /usr/lib/systemd/system/kibana.service
     perl -pi -e 's#ES_HOME=#ES_HOME='${INSTALL_PATH}'#g' /usr/lib/systemd/system/elasticsearch.service
     perl -pi -e 's#ES_PATH_CONF=#ES_PATH_CONF='${INSTALL_PATH}'#g' /usr/lib/systemd/system/elasticsearch.service
     perl -pi -e 's#ExecStart=#ExecStart='${INSTALL_PATH}'#g' /usr/lib/systemd/system/elasticsearch.service
    

31. Enable the system services

     systemctl daemon-reload
     systemctl reenable alert
     systemctl reenable kibana
     systemctl reenable elasticsearch
    

32. Set location for Elasticsearch data and logs files in configuration file

    • Elasticsearch

        perl -pi -e 's#path.data: #path.data: '${INSTALL_PATH}'#g' ${INSTALL_PATH}/etc/elasticsearch/elasticsearch.yml
        perl -pi -e 's#path.logs: #path.logs: '${INSTALL_PATH}'#g' ${INSTALL_PATH}/etc/elasticsearch/elasticsearch.yml
        perl -pi -e 's#/usr#'${INSTALL_PATH}'/usr#g' ${INSTALL_PATH}/etc/elasticsearch/jvm.options
        perl -pi -e 's#/usr#'${INSTALL_PATH}'/usr#g' /etc/sysconfig/elasticsearch
      

    • Kibana

        perl -pi -e 's#/etc#'${INSTALL_PATH}'/etc#g' ${INSTALL_PATH}/etc/kibana/kibana.yml
        perl -pi -e 's#/opt#'${INSTALL_PATH}'/opt#g' ${INSTALL_PATH}/etc/kibana/kibana.yml
        perl -pi -e 's#/usr#'${INSTALL_PATH}'/usr#g' ${INSTALL_PATH}/etc/kibana/kibana.yml
      

    • AI

        perl -pi -e 's#/opt#'${INSTALL_PATH}'/opt#g' ${INSTALL_PATH}/opt/ai/bin/conf.cfg
      

33. What next ?

    • Upload License file to ${INSTALL_PATH}/usr/share/elasticsearch/directory.

    • Setup cluster in ${INSTALL_PATH}/etc/elasticsearch/elasticsearch.yml

        discovery.zen.ping.unicast.hosts: [ "172.10.0.1:9300", "172.10.0.2:9300" ] 
      

    • Redirect GUI to 443/tcp

        	firewall-cmd --zone=public --add-masquerade --permanent
        	firewall-cmd --zone=public --add-forward-port=port=443:proto=tcp:toport=5601 --permanent
        	firewall-cmd --reload
      

Plugins management in the Elasticsearch

Base installation of the OP5 Log Analytics contains the elasticsearch-auth plugin. You can extend the basic Elasticsearch functionality by installing the custom plugins.

Plugins contain JAR files, but may also contain scripts and config files, and must be installed on every node in the cluster.

After installation, each node must be restarted before the plugin becomes visible.

The Elasticsearch provides two categories of plugins:

• Core Plugins - it is plugins that are part of the Elasticsearch project.
• Community contributed - it is plugins that are external to the Elasticsearch project

Installing Plugins

Core Elasticsearch plugins can be installed as follows:

cd /usr/share/elasticsearch/ 
bin/elasticsearch-plugin install [plugin_name]

Example:

bin/elasticsearch-plugin install ingest-geoip

-> Downloading ingest-geoip from elastic
[=================================================] 100%  
@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@
@ WARNING: plugin requires additional permissions @
@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@
* java.lang.RuntimePermission accessDeclaredMembers
* java.lang.reflect.ReflectPermission suppressAccessChecks
See http://docs.oracle.com/javase/8/docs/technotes/guides/security/permissions.html
for descriptions of what these permissions allow and the associated risks.

Continue with installation? [y/N]y
-> Installed ingest-geoip

Plugins from custom link or filesystem can be installed as follow:

cd /usr/share/elasticsearch/
sudo bin/elasticsearch-plugin install [url]

Example:

sudo bin/elasticsearch-plugin install file:///path/to/plugin.zip
bin\elasticsearch-plugin install file:///C:/path/to/plugin.zip
sudo bin/elasticsearch-plugin install http://some.domain/path/to/plugin.zip

Listing plugins

Listing currently loaded plugins

sudo bin/elasticsearch-plugin list

listing currently available core plugins:

sudo bin/elasticsearch-plugin list --help

Removing plugins

sudo bin/elasticsearch-plugin remove [pluginname]

Updating plugins

sudo bin/elasticsearch-plugin remove [pluginname]
sudo bin/elasticsearch-plugin install [pluginname]

Discovery

Time settings and refresh

In the upper right corner there is a section in which it defines the range of time that OP5 will search in terms of conditions contained in the search bar. The default value is the last 15 minutes.

After clicking this selection, we can adjust the scope of search by selecting one of the three tabs in the drop-down window:

• Quick: contain several predefined ranges that should be clicked.
• Relative: in this windows specify the day from which OP5 Log Analytics should search for data.
• Absolute: using two calendars we define the time range for which the search results are to be returned.

Fields

OP5 Log Analytics in the body of searched events, it recognize fields that can be used to created more precision queries. The extracted fields are visible in the left panel. They are divided on three types: timestamp, marked on clock icon ; text, marked with the letter “t” and digital, marked witch hashtag .

Pointing to them and clicking on icon , they are automatically transferred to the „Selected Fields” column and in the place of events a table with selected columns is created on regular basis. In the “Selected Fields” selection you can also delete specific fields from the table by clicking on the selected element.

Filtering and syntax building

We use the query bar to search interesting events. For example, after entering the word „error”, all events that contain the word will be displayed, additional highlighting them with an yellow background.

Syntax

Fields can be used in the similar way by defining conditions that interesting us. The syntax of such queries is:

<fields_name:<fields_value>

Example:

status:500

This query will display all events that contain the „status” fields with a value of 500.

Filters

The field value does not have to be a single, specific value. For digital fields we can specify range in the following scheme:

<fields_name:[<range_from TO <range_to] 

Example:

status:[500 TO 599]

This query will return events with status fields that are in the range 500 to 599.

Operators

The search language used in OP5 allows to you use logical operators „AND”, „OR” and „NOT”, which are key and necessary to build more complex queries.

• AND is used to combined expressions, e.g. „error AND „access denied”. If an event contain only one expression or the words error and denied but not the word access, then it will not be displayed.
• OR is used to search for the events that contain one OR other expression, e.g. „status:500” OR “denied”. This query will display events that contain word „denied” or status field value of 500. OP5 uses this operator by default, so query „status:500” “denied” would return the same results.
• NOT is used to exclude the following expression e.g. „status:[500 TO 599] NOT status:505” will display all events that have a status fields, and the value of the field is between 500 and 599 but will eliminate from the result events whose status field value is exactly 505.
• The above methods can be combined with each other by building even more complex queries. Understanding how they work and joining it, is the basis for effective searching and full use of OP5 Log Analytics.

Example of query built from connected logical operations:

status:[500 TO 599] AND („access denied" OR error) NOT status:505

Returns in the results all events for which the value of status fields are in the range of 500 to 599, simultaneously contain the word „access denied” or „error”, omitting those events for which the status field value is 505.

Saving and deleting queries

Saving queries enables you to reload and use them in the future.

Save query

To save query, click on the “Save” button under on the query bar:

This will bring up a window in which we give the query a name and then click the button .

Saved queries can be opened by going to „Open” from the main menu at the top of the page, and select saved search from the search list:

Additional you can use “Saved Searchers Filter..” to filter the search list.

Open query

To open a saved query from the search list, you can click on the name of the query you are interested in.

After clicking on the icon on the name of the saved query and chose “Edit Query DSL”, we will gain access to the advanced editing mode, so that we can change the query on at a lower level.

It is a powerful tool designed for advanced users, designed to modify the query and the way it is presented by OP5 Log Analytics.

Delete query

To delete a saved query, open it from the search list, and then click on the button .

If you want delete many saved queries simultaneously go to the “Management Object” -> “Saved Object” -> “Searches” select it in the list (the icon to the left of the query name), and then click “Delete” button.

From this level, you can also export saved queries in the same way. To do this, you need to click on and choose the save location. The file will be saved in .JSON format. If you then want to import such a file to OP5 Log Analytics, click on button , at the top of the page and select the desired file.

Visualizations

Visualize enables you to create visualizations of the data in your OP5 Log Analytics indices. You can then build dashboards that display related visualizations. Visualizations are based on OP5 Log Analytics queries. By using a series of OP5 Log Analytics aggregations to extract and process your data, you can create charts that show you the trends, spikes, and dips.

Creating visualization

Create

To create visualization, go to the „Visualize” tab from the main menu. A new page will be appearing where you can create or load visualization.

Load

To load previously created and saved visualization, you must select it from the list.

In order to create a new visualization, you should choose the preferred method of data presentation.

Next, specify whether the created visualization will be based on a new or previously saved query. If on new one, select the index whose visualization should concern. If visualization is created from a saved query, you just need to select the appropriate query from the list, or (if there are many saved searches) search for them by name.

Vizualization types

Before the data visualization will be created, first you have to choose the presentation method from an existing list. Currently there are five groups of visualization types. Each of them serves different purposes. If you want to see only the current number of products sold, it is best to choose „Metric”, which presents one value.

However, if we would like to see user activity trends on pages in different hour and days, a better choice will be „Area chart”, which displays a chart with time division.

The „Markdown widget” views is used to place text e.g. information about the dashboard, explanations and instruction on how to navigate. Markdown language was used to format the text (the most popular use is GitHub). More information and instruction can be found at this link: https://help.github.com/categories/writing-on-github/

Edit visualization and saving

Edititing

Editing a saved visualization enables you to directly modify the object definition. You can change the object title, add a description, and modify the JSON that defines the object properties. After selecting the index and the method of data presentation, you can enter the editing mode. This will open a new window with empty visualization.

At the very top there is a bar of queries that cat be edited throughout the creation of the visualization. It work in the same way as in the “Discover” tab, which means searching the raw data, but instead of the data being displayed, the visualization will be edited. The following example will be based on the „Area chart”. The visualization modification panel on the left is divided into three tabs: „Data”, “Metric & Axes” and „Panel Settings”.

In the „Data” tab, you can modify the elements responsible for which data and how should be presented. In this tab there are two sectors: “metrics”, in which we set what data should be displayed, and „buckets” in which we specify how they should be presented.

Select the Metrics & Axes tab to change the way each individual metric is shown on the chart. The data series are styled in the Metrics section, while the axes are styled in the X and Y axis sections.

In the „Panel Settings” tab, there are settings relating mainly to visual aesthetics. Each type of visualization has separate options.

To create the first graph in the char modification panel, in the „Data” tab we add X-Axis in the “buckets” sections. In „Aggregation” choose „Histogram”, in „Field” should automatically be located “timestamp” and “interval”: “Auto” (if not, this is how we set it). Click on the icon on the panel. Now our first graph should show up.

Some of the options for „Area Chart” are:

Smooth Lines - is used to smooth the graph line.

• Current time marker – places a vertical line on the graph that determines the current time.

• Set Y-Axis Extents – allows you to set minimum and maximum values for the Y axis, which increases the readability of the graphs. This is useful, if we know that the data will never be less then (the minimum value), or to indicate the goals the company (maximum value).

• Show Tooltip – option for displaying the information window under the mouse cursor, after pointing to the point on the graph.

  

Saving

To save the visualization, click on the “Save” button under on the query bar: give it a name and click the button .

Load

To load the visualization, go to the “Management Object” -> “Saved Object” -> “Visualizations” select it from the list. From this place, we can also go into advanced editing mode. To view of the visualization use button.# Dashboards #

Dashboards

Dashboard is a collection of several visualizations or searches. Depending on how it is build and what visualization it contains, it can be designed for different teams e.g.:

• SOC - which is responsible for detecting failures or threats in the company;
• business - which thanks to the listings can determine the popularity of products and define the strategy of future sales and promotions;
• managers and directors - who may immediately have access to information about the performance units or branches.

Create

To create a dashboard from previously saved visualization and queries, go to the „Dashboard” tab in the main menu. When you open it, a new page will appear.

Clicking on the icon “Add” at the top of page select “Visualization” or “Saved Search” tab.

and selecting a saved query and / or visualization from the list will add them to the dashboard. If, there are a large number of saved objects, use the bar to search for them by name.

Elements of the dashboard can be enlarged arbitrarily (by clicking on the right bottom corner of object and dragging the border) and moving (by clicking on the title bar of the object and moving it).

Saving

To save a dashboard, click on the “Save” button to the up of the query bar and give it a name.

Load

To load the Dashboard, go to the “Management Object” -> “Saved Object” -> “Dashborad” select it from the list. From this place, we can also go into advanced editing mode. To view of the visualization use button.# Sharing dashboards #

Sharing dashboards

The dashboard can be share with other OP5 Log Analytics users as well as on any page - by placing a snippet of code. Provided that it cans retrieve information from OP5 Log Analytics.

To do this, create new dashboard or open the saved dashboard and click on the “Share” to the top of the page. A window will appear with generated two URL. The content of the first one “Embaded iframe” is used to provide the dashboard in the page code, and the second “Link” is a link that can be passed on to another user. There are two option for each, the first is to shorten the length of the link, and second on copies to clipboard the contest of the given bar.

Dashboard drilldown

In discovery tab search for message of Your interest

---

Save Your search

Check You „Shared link” and copy it

! ATTENTION ! Do not copy „?_g=()” at the end.

---

Select Alerting module

---

Once Alert is created use ANY frame to add the following directives:

Use_kibana4_dashboard: paste Your „shared link” here

use_kibana_dashboard: - The name of a Kibana dashboard to link to. Instead of generating a dashboard from a template, Alert can use an existing dashboard. It will set the time range on the dashboard to around the match time, upload it as a temporary dashboard, add a filter to the query_key of the alert if applicable, and put the url to the dashboard in the alert. (Optional, string, no default).

---

Kibana4_start_timedelta

kibana4_start_timedelta: Defaults to 10 minutes. This option allows you to specify the start time for the generated kibana4 dashboard. This value is added in front of the event. For example,

`kibana4_start_timedelta: minutes: 2`

---

Kibana4_end_timedelta`

kibana4_end_timedelta: Defaults to 10 minutes. This option allows you to specify the end time for the generated kibana4 dashboard. This value is added in back of the event. For example,

kibana4_end_timedelta: minutes: 2

---

Sample:

---

Search for triggered alert in Discovery tab. Use alert* search pattern.

Refresh the alert that should contain url for the dashboard. Once available, kibana_dashboard field can be exposed to dashboards giving You a real drill down feature.

Reports

OP5 Log Analytics contains a module for creating reports that can be run cyclically and contain only interesting data, e.g. a weekly sales report.

To go to the reports windows, select to tiles icon from the main menu bar, and then go to the „Reports” icon (To go back, go to the „Search” icon).

CSV Report

To export data to CSV Report click the Reports icon, you immediately go to the first tab - Export Task Management.

In this tab we have the opportunity to specify the source from which we want to do export. It can be an index pattern. After selecting it, we confirm the selection with the Submit button and a report is created at the moment. The symbol can refresh the list of reports and see what its status is.

We can also create a report by pointing to a specific index from the drop-down list of indexes.

We can also check which fields are to be included in the report. The selection is confirmed by the Submit button.

When the process of generating the report (Status:Completed) is finished, we can download it (Download button) or delete (Delete button). The downloaded report in the form of *.csv file can be opened in the browser or saved to the disk.

In this tab, the downloaded data has a format that we can import into other systems for further analysis.

PDF Report

In the Export Dashboard tab we have the possibility to create graphic reports in PDF files. To create such a report, just from the drop-down list of previously created and saved Dashboards, indicate the one we are interested in, and then confirm the selection with the Submit button. A newly created export with the Processing status will appear on the list under Dashboard Name. When the processing is completed, the Status changes to Complete and it will be possible to download the report.

By clicking the Download button, the report is downloaded to the disk or we can open it in the PDF file browser. There is also to option of deleting the report with the Delete button.

Below is an example report from the Dashboard template generated and downloaded as a PDF file.

Scheduler Report (Schedule Export Dashboard)

In the Report selection, we have the option of setting the Scheduler which from Dashboard template can generate a report at time intervals. To do this goes to the Schedule Export Dashboard tab.

In this tab mark the saved Dashboard.

In the Email Topic field, enter the Message title, in the Email field enter the email address to which the report should be sent. From drop-down list choose at what frequency you want the report to be generated and sent. The action configured in this way is confirmed with the Submit button.

The defined action goes to the list and will generate a report to the e-mail address, with the cycle we set, until we cannot cancel it with the Cancel button.

User roles and object management

Users, roles and settings

OP5 Log Analytics allows to you manage users and permission for indexes and methods used by them. To do this click the “Config” button from the main menu bar.

A new window will appear with three main tabs: „User Management”, „Settings” and „License Info”.

From the „User Management” level we have access to the following possibilities: Creating a user in „Create User”, displaying users in „User List”, creating new roles in „Create roles” and displaying existing roles in „List Role”.

Creating a User (Create User)

Creating user

To create a new user click on the Config icon and you immediately enter the administration panel, where the first tab is to create a new user (Create User).

In the wizard that opens, we enter a unique username (Username field), password for the user (field Password) and assign a role (field Role). In this field we have the option of assigning more than one role. Until we select role in the Roles field, the Default Role field remains empty. When we mark several roles, these roles appear in the Default Role field. In this field we have the opportunity to indicate which role for a new user will be the default role with which the user will be associated in the first place when logging in. The default role field has one more important task - it binds all users with the field / role set in one group. When one of the users of this group create Visualization or Dashboard it will be available to other users from this role(group). Creating the account is confirmed with the Submit button.

User’s modification and deletion, (User List)

Once we have created users, we can display their list. We do it in next tab (User List).

In this view, we get a list of user account with assigned roles and we have two buttons: Delete and Update. The first of these is ability to delete a user account. Under the Update button is a drop-down menu in which we can change the previous password to a new one (New password), change the password (Re-enter Ne Password), change the previously assigned roles (Roles), to other (we can take the role assigned earlier and give a new one, extend user permissions with new roles). The introduced changes are confirmed with the Submit button.

We can also see current user setting and clicking the Update button collapses the previously expanded menu.

Create, modify and delete a role (Create Role), (Role List)

In the Create Role tab we can define a new role with permissions that we assign to a pattern or several index patterns.

In example, we use the syslog2* index pattern. We give this name in the Paths field. We can provide one or more index patterns, their names should be separated by a comma. In the next Methods field, we select one or many methods that will be assigned to the role. Available methods:

• PUT - sends data to the server
• POST - sends a request to the server for a change
• DELETE - deletes the index / document
• GET - gets information about the index /document
• HEAD - is used to check if the index /document exists

In the role field, enter the unique name of the role. We confirm addition of a new role with the Submit button. To see if a new role has been added, go to the net Role List tab.

As we can see, the new role has been added to the list. With the Delete button we have the option of deleting it, while under the Update button we have a drop-down menu thanks to which we can add or remove an index pattern and add or remove a method. When we want to confirm the changes, we choose the Submit button. Pressing the Update button again will close the menu.

Fresh installation of the application have sewn solid roles which granting user special rights:

• admin - this role gives unlimited permissions to administer / manage the application
• alert - a role for users who want to see the Alert module
• kibana - a role for users who want to see the application GUI
• Intelligence - a role for users who are to see the Intelligence moduleObject access permissions (Objects permissions)

---

In the User Manager tab we can parameterize access to the newly created role as well as existing roles. In this tab we can indicate to which object in the application the role has access.

Example:

In the Role List tab we have a role called sys2, it refers to all index patterns beginning with syslog* and the methods get, post, delete, put and head are assigned.

When we go to the Object permission tab, we have the option to choose the sys2 role in the drop-down list choose a role:

After selecting, we can see that we already have access to the objects: two index patterns syslog2* and op5-* and on dashboard Windows Events. There are also appropriate read or updates permissions.

From the list we have the opportunity to choose another object that we can add to the role. We have the ability to quickly find this object in the search engine (Find) and narrowing the object class in the drop-down field “Select object type”. The object type are associated with saved previously documents in the sections Dashboard, Index pattern, Search and Visualization. By buttons we have the ability to add or remove or object, and Save button to save the selection.

Default user and passwords

The table below contains built-in user accounts and default passwords:

|Address                |User         |Password     |Description                                     |Usage          |
|-----------------------|-------------|-------------|------------------------------------------------|---------------|
|https://localhost:5601	|logserver    |logserver    |A built-in *superuser* account                  |               |
|                       |alert        |alert        |A built-in account for  the Alert module        |               |  
|                       |intelligence |intelligece  |A built-in account for the Intelligence module  | authorizing communication with elasticsearch server | 
|                       |scheduler    |scheduler    |A built-in account for the Scheduler module     |
|                       |logstash     |logstash     |A built-in account for authorized comuunication form Logstash |

Changing password for the system account

After you change password for one of the system account ( alert, intelligence, logserver, scheduler), you must to do appropriate changes in the application files.

1. Account Logserver

   • Update /etc/kibana/kibana.yml:

       vi /etc/kibana/kibana.yml
     
       elasticsearch.password: new_logserver_passowrd
       elastfilter.password: "new_logserver_password"
     

2. Account Intelligence

   • Update /opt/ai/bin/conf.cfg

       vi /opt/ai/bin/conf.cfg
       password=new_intelligence_password
     

3. Account Alert

   • Update file /opt/alert/config.yaml

       vi /opt/alert/config.yaml
       es_password: alert
     

4. Account Scheduler

   • Update /etc/kibana/kibana.yml:

       vi /etc/kibana/kibana.yml	
       elastscheduler.password: "new_scheduler_password"
     

Settings

General Settings

The Settings tab is used to set the audit on different activates or events and consists of several fields:

• Time Out in minutes field - this field defines the time after how many minutes the application will automatically log you off
• Delete Application Tokens (in days) - in this field we specify after what time the data from the audit should be deleted
• Delete Audit Data (in days) field - in this field we specify after what time the data from the audit should be deleted
• Next field are checkboxes in which we specify what kind of events are to be logged (saved) in the audit index. The events that can be monitored are: logging (Login), logging out (Logout), creating a user (Create User), deleting a user (Delete User), updating user (Update User), creating a role (Create Role), deleting a role (Delete Role), update of the role (Update Role), start of export (Export Start), delete of export (Export Delete), queries (Queries), result of the query (Content), if attempt was made to perform a series of operation (Bulk)
• Delete Exported CSVs (in days) field - in this field we specify after which time exported file with CSV extension have to be removed
• Delete Exported PDFs (in days) field - in this field we specify after which time exported file with PDF extension have to be removed

To each field is assigned “Submit” button thanks to which we can confirm the changes.

License (License Info)

The License Information tab consists of several non-editable information fields.

These fields contain information:

• Company field, who owns the license - in this case EMCA S.A.
• Data nodes in cluster field - how many nodes we can put in one cluster - in this case 100
• No of documents field - empty field
• Indices field - number of indexes, symbol[*] means that we can create any number of indices
• Issued on field - date of issue
• Validity field - validity, in this case for 360000 months

Special accounts

At the first installation of the OP5 Log Analytics application, apart from the administrative account (logserver), special applications are created in the application: alert, intelligence and scheduler.

• Alert Account - this account is connected to the Alert Module which is designed to track events written to the index for the previously defined parameters. If these are met the information action is started (more on the action in the Alert section)
• Intelligence Account - with this account is related to the module of artificial intelligence which is designed to track events and learn the network based on previously defined rules artificial intelligence based on one of the available algorithms (more on operation in the Intelligence chapter)
• Scheduler Account - the scheduler module is associated with this account, which corresponds to, among others for generating reports

Alert Module

OP5 Log Analytics allows you to create alerts, i.e. monitoring queries. These are constant queries that run in the background and when the conditions specified in the alert are met, the specify action is taken.

For example, if you want to know when more than 20 „status:500” responscode from on our homepage appear within an one hour, then we create an alert that check the number of occurrences of the „status:500” query for a specific index every 5 minutes. If the condition we are interested in is met, we send an action in the form of sending a message to our e-mail address. In the action, you can also set the launch of any script.

Enabling the Alert Module

To enabling the alert module you should:

• generate writeback index for Alert service:

    /opt/alert/bin/elastalert-create-index --config /opt/alert/config.yaml
  

• configure the index pattern for alert*

• start the Alert service:

    systemctl start alert
  

Creating Alerts

To create the alert, click the “Alerts” button from the main menu bar.

We will display a page with tree tabs: Create new alerts in „Create alert rule”, manage alerts in „Alert rules List” and check alert status „Alert Status”.

In the alert creation windows we have an alert creation form:

• Name - the name of the alert, after which we will recognize and search for it.
• Index pattern - a pattern of indexes after which the alert will be searched.
• Role - the role of the user for whom an alert will be available
• Type - type of alert
• Description - description of the alert.
• Example - an example of using a given type of alert. Descriptive field
• Alert method - the action the alert will take if the conditions are met (sending an email message or executing a command)
• Any - additional descriptive field.# List of Alert rules #

The “Alert Rule List” tab contain complete list of previously created alert rules:

In this window, you can activate / deactivate, delete and update alerts by clicking on the selected icon with the given alert: .

Alerts status

In the “Alert status” tab, you can check the current alert status: if it activated, when it started and when it ended, how long it lasted, how many event sit found and how many times it worked.

Also, on this tab, you can recover the alert dashboard, by clicking the “Recovery Alert Dashboard” button.# Type of the Alert module rules #

The various RuleType classes, defined in OP5-Log-Aalytics. An instance is held in memory for each rule, passed all of the data returned by querying Elasticsearch with a given filter, and generates matches based on that data.

• Any - The any rule will match everything. Every hit that the query returns will generate an alert.
• Blacklist - The blacklist rule will check a certain field against a blacklist, and match if it is in the blacklist.
• Whitelist - Similar to blacklist, this rule will compare a certain field to a whitelist, and match if the list does not contain the term.
• Change - This rule will monitor a certain field and match if that field changes.
• Frequency - his rule matches when there are at least a certain number of events in a given time frame.
• Spike - This rule matches when the volume of events during a given time period is spike_height times larger or smaller than during the previous time period.
• Flatline - This rule matches when the total number of events is under a given threshold for a time period.
• New Term - This rule matches when a new value appears in a field that has never been seen before.
• Cardinality - This rule matches when a the total number of unique values for a certain field within a time frame is higher or lower than a threshold.
• Metric Aggregation - This rule matches when the value of a metric within the calculation window is higher or lower than a threshold.
• Percentage Match - This rule matches when the percentage of document in the match bucket within a calculation window is higher or lower than a threshold.

Example of rules

Unix - Authentication Fail

• index pattern:

    syslog-*
  

• Type:

    Frequency
  

• Alert Method:

    Email
  

• Any:

    num_events: 4
    timeframe:
      minutes: 5
  
    filter:
    - query_string:
        query: "program: (ssh OR sshd OR su OR sudo) AND message: \"Failed password\""
  

Windows - Firewall disable or modify

• index pattern:

    beats-*
  

• Type:

    Any
  

• Alert Method:

    Email
  

• Any:

filter:

	- query_string:
	       query: "event_id:(4947 OR 4948 OR 4946 OR 4949 OR 4954 OR 4956 OR 5025)"

Playbooks

OP5 Log Analytics has a set of predefined set of rules and activities (called Playbook) that can be attached to a registered event in the Alert module. Playbooks can be enriched with scripts that can be launched together with Playbook.

Create Playbook

To add a new playbook, go to the Alert module, select the Playbook tab and then Create Playbook

In the Name field, enter the name of the new Playbook.

In the Text field, enter the content of the Playbook message.

In the Script field, enter the commands to be executed in the script.

To save the entered content, confirm with the Submit button.

Playbooks list

To view saved Playbook, go to the Alert module, select the Playbook tab and then Playbooks list:

To view the content of a given Playbook, select the Show button.

To enter the changes in a given Playbook or in its script, select the Update button. After making changes, select the Submit button.

To delete the selected Playbook, select the Delete button.

Linking Playbooks with alert rule

You can add a Playbook to the Alert while creating a new Alert or by editing a previously created Alert.

To add Palybook to the new Alert rule, go to the Create alert rule tab and in the Playbooks section use the arrow keys to move the correct Playbook to the right window.

To add a Palybook to existing Alert rule, go to the Alert rule list tab with the correct rule select the Update button and in the Playbooks section use the arrow keys to move the correct Playbook to the right window.

Playbook verification

When creating an alert or while editing an existing alert, it is possible that the system will indicate the most-suited playbook for the alert. For this purpose, the Validate button is used, which starts the process of searching the existing playbook and selects the most appropriate ones.

Risks

OP5 Log Analytics allows you to estimate the risk based on the collected data. The risk is estimated based on the defined category to which the values from 0 to 100 are assigned.

Information on the defined risk for a given field is passed with an alert and multiplied by the value of the Rule Importance parameter.

Create category

To add a new risk Category, go to the Alert module, select the Risks tab and then Create Cagtegory.

Enter the Name for the new category and the category Value.

Category list

To view saved Category, go to the Alert module, select the Risks tab and then Categories list:

To view the content of a given Category, select the Show button.

To change the value assigned to a category, select the Update button. After making changes, select the Submit button.

To delete the selected Category, select the Delete button.

Create risk

To add a new playbook, go to the Alert module, select the Playbook tab and then Create Playbook

In the Index pattern field, enter the name of the index pattern. Select the Read fields button to get a list of fields from the index. From the box below, select the field name for which the risk will be determined.

From the Timerange field, select the time range from which the data will be analyzed.

Press the Read valules button to get values from the previously selected field for analysis.

Next, you must assign a risk category to the displayed values. You can do this for each value individually or use the check-box on the left to mark several values and set the category globally using the Set global category button. To quickly find the right value, you can use the search field.

After completing, save the changes with the Submit button.

List risk

To view saved risks, go to the Alert module, select the Risks tab and then Risks list:

To view the content of a given Risk, select the Show button.

To enter the changes in a given Risk, select the Update button. After making changes, select the Submit button.

To delete the selected Risk, select the Delete button.

Linking risk with alert rule

You can add a Risk key to the Alert while creating a new Alert or by editing a previously created Alert.

To add Risk key to the new Alert rule, go to the Create alert rule tab and after entering the index name, select the Read fields button and in the Risk key field, select the appropriate field name. In addition, you can enter the validity of the rule in the Rule Importance field (in the range 1-100%), by which the risk will be multiplied.

To add Risk key to the existing Alert rule, go to the Alert rule list, tab with the correct rule select the Update button. Use the Read fields button and in the Risk key field, select the appropriate field name. In addition, you can enter the validity of the rule in the Rule Importance field (in the range 1-100%), by which the risk will be multiplied.

Risk calculation algorithms

The risk calculation mechanism performs the aggregation of the risk field values. We have the following algorithms for calculating the alert risk (Aggregation type):

• min - returns the minimum value of the risk values from selected fields;
• max - returns the maximum value of the risk values from selected fields;
• avg - returns the average of risk values from selected fields;
• sum - returns the sum of risk values from selected fields;
• custom - returns the risk value based on your own algorithm

Adding a new risk calculation algorithm

The new algorithm should be added in the ./elastalert_modules/playbook_util.py file in the calculate_risk method. There is a sequence of conditional statements for already defined algorithms:

#aggregate values by risk_key_aggregation for rule
if risk_key_aggregation == "MIN":
    value_agg = min(values)
elif risk_key_aggregation == "MAX":
    value_agg = max(values)
elif risk_key_aggregation == "SUM":
    value_agg = sum(values)
elif risk_key_aggregation == "AVG":
    value_agg = sum(values)/len(values)
else:
    value_agg = max(values)

To add a new algorithm, add a new sequence as shown in the above code:

elif risk_key_aggregation == "AVG":
    value_agg = sum(values)/len(values)
elif risk_key_aggregation == "AAA":
    value_agg = BBB
else:
    value_agg = max(values)

where AAA is the algorithm code, BBB is a risk calculation function.

Using the new algorithm

After adding a new algorithm, it is available in the GUI in the Alert tab.

To use it, add a new rule according to the following steps:

• Select the custom value in the Aggregation type field;
• Enter the appropriate value in the Any field, e.g. risk_key_aggregation: AAA

The following figure shows the places where you can call your own algorithm:

Additional modification of the algorithm (weight)

Below is the code in the calcuate_risk method where category values are retrieved - here you can add your weight:

        #start loop by tablicy risk_key
        for k in range(len(risk_keys)):
            risk_key = risk_keys[k]
            logging.info(' >>>>>>>>>>>>>> risk_key: ')
            logging.info(risk_key)
            key_value = lookup_es_key(match, risk_key)
            logging.info(' >>>>>>>>>>>>>> key_value: ')
            logging.info(key_value)
            value = float(self.get_risk_category_value(risk_key, key_value))
            values.append( value )
            logging.info(' >>>>>>>>>>>>>> risk_key values: ')
            logging.info(values)
        #finish loop by tablicy risk_key
        #aggregate values by risk_key_aggregation form rule
        if risk_key_aggregation == "MIN":
            value_agg = min(values)
        elif risk_key_aggregation == "MAX":
            value_agg = max(values)
        elif risk_key_aggregation == "SUM":
            value_agg = sum(values)
        elif risk_key_aggregation == "AVG":
            value_agg = sum(values)/len(values)
        else:
            value_agg = max(values)

Risk_key is the array of selected risk key fields in the GUI. A loop is made on this array and a value is collected for the categories in the line:

value = float(self.get_risk_category_value(risk_key, key_value))

Based on, for example, Risk_key, you can multiply the value of the value field by the appropriate weight. The value field value is then added to the table on which the risk calculation algorithms are executed.

Intelligence Module

A dedicated artificial intelligence module has been built in the OP5 Log Analytics system that allows prediction of parameter values relevant to the maintenance of infrastructure and IT systems. Such parameters include:

• use of disk resources,
• use of network resources,
• using the power of processors
• detection of known incorrect behaviour of IT systems

To access of the Intelligence module, click the tile icon from the main meu bar and then go to the „Intelligence” icon (To go back, click to the „Search” icon).

There are 4 screens available in the module:

• Create AI Rule - the screen allows you to create artificial intelligence rules and run them in scheduler mode or immediately
• AI Rules List - the screen presents a list of created artificial intelligence rules with the option of editing, previewing and deleting them
• AI Learn - the screen allows to define the conditions for teaching the MLP neural network
• AI Learn Tasks - a screen on which the initiated and completed learning processes of neural networks with the ability to preview learning results are presented.# Create AI Rule #

To create the AI Rule, click on the tile icon from the main menu bar, go to the „Intelligence” icon and select “Create AI Rule” tab. The screen allows to defining the rules of artificial intelligence based on one of the available algorithms (a detailed description of the available algorithms is available in a separate document).

Description of the controls available on the fixed part of screen:

• Algorithm - the name of the algorithm that forms the basis of the artificial intelligence rule
• Choose search - search defined in the OP5 Log Analytics system, which is used to select a set of data on which the artificial intelligence rule will operate
• Run - a button that allows running the defined AI rule or saving it to the scheduler and run as planned

The rest of the screen will depend on the chosen artificial intelligence algorithm.

The fixed part of the screen

Description of the controls available on the fixed part of screen:

• Algorithm - the name of the algorithm that forms the basis of the artificial intelligence rule
• Choose search - search defined in the OP5 Log Analytics system, which is used to select a set of data on which the artificial intelligence rule will operate
• Run - a button that allows running the defined AI rule or saving it to the scheduler and run as planned

The rest of the screen will depend on the chosen artificial intelligence algorithm.

Screen content for regressive algorithms

Description of controls:

• feature to analyze from search - analyzed feature (dictated)
• multiply by field - enable multiplication of algorithms after unique values of the feature indicated here. Multiplication allows you to run the AI rule one for e.g. all servers. The value “none” in this field means no multiplication.
• multiply by values - if a trait is indicated in the „multiply by field”, then unique values of this trait will appear in this field. Multiplications will be made for the selected values. If at least one of value is not selected, the „Run” buttons will be inactive.`

In other words, multiplication means performing an analysis for many values from the indicated field, for example: sourece_node_host- which we indicate in Multiply by field (from search).

However, in Multiply by values (from search) we already indicate values of this field for which the analysis will be performed, for example: host1, host2, host3, ….

• time frame - feature aggregation method (1 minute, 5 minute, 15 minute, 30 minute, hourly, weekly, monthly, 6 months, 12 months)
• max probes - how many samples back will be taken into account for analysis. A single sample is an aggregated data according to the aggregation method.
• value type - which values to take into account when aggregating for a given time frame (e.g. maximum from time frame, minimum, average)
• max predictions - how many estimates we make for ahead (we take time frame)
• data limit - limits the amount of date downloaded from the source. It speeds up processing but reduces its quality
• start date - you can set a date earlier than the current date in order to verify how the selected algorithm would work on historical data
• Scheduler - a tag if the rule should be run according to the plan for the scheduler. If selected, additional fields will appear;

• Prediction cycle - plan definition for the scheduler, i.e. the cycle in which the prediction rule is run (e.g. once a day, every hour, once a week). In the field, enter the command that complies with the cron standard. Enable – whether to immediately launch the scheduler plan or save only the definition
• Role - only users with the roles selected here and the administrator will be able to run the defend AI rules The selected „time frame” also affects the prediction period. If we choose “time frame = monthly”, we will be able to predict a one month ahead from the moment of prediction (according to the “prediction cycle” value)

Screen content for the Trend algorithm

Description of controls:

• feature to analyze from search - analyzed feature (dictated)
• multiply by field - enable multiplication of algorithms after unique values of the feature indicated here. Multiplication allows you to run the AI rule one for e.g. all servers. The value “none” in this field means no multiplication.
• multiply by values - if a trait is indicated in the „multiply by field”, then unique values of this trait will appear in this field. Multiplications will be made for the selected values. If at least one of value is not selected, the „Run” buttons will be inactive.`

In other words, multiplication means performing an analysis for many values from the indicated field, for example: sourece_node_host- which we indicate in Multiply by field (from search).

However, in Multiply by values (from search) we already indicate values of this field for which the analysis will be performed, for example: host1, host2, host3, ….

• time frame - feature aggregation method (1 minute, 5 minute, 15 minute, 30 minute, hourly, weekly, monthly, 6 months, 12 months)
• max probes - how many samples back will be taken into account for analysis. A single sample is an aggregated data according to the aggregation method.
• value type - which values to take into account when aggregating for a given time frame (e.g. maximum from time frame, minimum, average)
• max predictions - how many estimates we make for ahead (we take time frame)
• data limit - limits the amount of date downloaded from the source. It speeds up processing but reduces its quality
• start date - you can set a date earlier than the current date in order to verify how the selected algorithm would work on historical data
• Scheduler - a tag if the rule should be run according to the plan for the scheduler. If selected, additional fields will appear;

• Prediction cycle - plan definition for the scheduler, i.e. the cycle in which the prediction rule is run (e.g. once a day, every hour, once a week). In the field, enter the command that complies with the cron standard. Enable – whether to immediately launch the scheduler plan or save only the definition
• Role - only users with the roles selected here and the administrator will be able to run the defend AI rules The selected „time frame” also affects the prediction period. If we choose “time frame = monthly”, we will be able to predict a one month ahead from the moment of prediction (according to the “prediction cycle” value)
• Threshold - default values -1 (do not search). Specifies the algorithm what level of exceeding the value of the feature „feature to analyze from cheese” is to look for. The parameter currently used only by the “Trend” algorithm.

Screen content for the neural network (MLP) algorithm

Descriptions of controls:

• Name - name of the learned neural network
• Choose search - search defined in OP5 Log Analytics, which is used to select a set of data on which the rule of artificial intelligence will work
• Below, on the left, a list of attributes and their weights based on teaching ANN will be defined during the teaching. The user for each attribute will be able to indicate the field from the above mentioned search, which contain the values of the attribute and which will be analyzed in the algorithm. The presented list (for input and output attributes) will have a static and dynamic part. Static creation by presenting key with the highest weights. The key will be presented in the original form, i.e. perf_data./ The second part is a DropDown type list that will serve as a key update according to the user’s naming. On the right side, the attribute will be examined in a given rule / pattern. Here also the user must indicate a specific field from the search. In both cases, the input and output are narrowed based on the search fields indicated in Choose search.
• Data limit - limits the amount of data downloaded from the source. It speeds up the processing, but reduces its quality.
• Scheduler - a tag if the rule should be run according to the plan or the scheduler. If selected, additional fields will appear:

• Prediction cycle - plan definition for the scheduler, i.e. the cycle in which the prediction rule is run (e.g. once a day, every hour, once a week). In the field, enter the command that complies with the cron standard
• Enable - whether to immediately launch the scheduler plan or save only the definition
• Role - only users with the roles selected here and the administrator will be able to run the defined AI rules

AI Rules List

Column description:

• Status:
  • - the process is being processed (the pid of the process is in brackets)
  • - process completed correctly
  • - the process ended with an error
• Name - the name of the rule
• Search - the search on which the rule was run
• Method - an algorithm used in the AI rule
• Actions - allowed actions:
  • Show - preview of the rule definition
  • Enable/Disable - rule activation /deactivation
  • Delete - deleting the rule
  • Update - update of the rule definition
  • Preview - preview of the prediction results (the action is available after the processing has been completed correctly).

AI Learn

Description of controls:

• Search - a source of data for teaching the network
• prefix name - a prefix added to the id of the learned model that allows the user to recognize the model
• Input cols - list of fields that are analyzed / input features. Here, the column that will be selected in the output col should not be indicated. Only those columns that are related to processing should be selected. **
• Output col - result field, the recognition of which is learned by the network. This field should exist in the learning and testing data, but in the production data is unnecessary and should not occur. This field cannot be on the list of selected fields in “input col”.
• Output class category - here you can enter a condition in SQL format to limit the number of output categories e.g. if((outputCol) \< 10,(floor((outputCol))+1), Double(10)). This condition limits the number of output categories to 10. Such conditions are necessary for fields selected in “output col” that have continuous values. They must necessarily by divided into categories. In the Condition, use your own outputCol name instead of the field name from the index that points to the value of the “output col” attribute.
• Time frame - a method of aggregation of features to improve their quality (e.g. 1 minute, 5 minutes, 15 minutes, 30 minutes, 1 hour, 1 daily).
• Time frames output shift - indicates how many time frame units to move the output category. This allows teaching the network with current attributes, but for categories for the future.
• Value type - which values to take into account when aggregating for a given time frame (e.g. maximum from time frame, minimum, average)
• Output class count- the expected number of result classes. If during learning the network identifies more classes than the user entered, the process will be interrupted with an error, therefore it is better to set up more classes than less, but you have to keep in mind that this number affects the learning time.
• Neurons in first hidden layer (from, to) - the number of neurons in the first hidden layer. Must have a value > 0. Jump every 1.
• Neurons in second hidden layer (from, to) - the number of neurons in second hidden layer. If = 0, then this layer is missing. Jump every 1.
• Neurons in third hidden layer (from, to) - the number of neurons in third hidden layer. If = 0 then this layer is missing. Jump every 1.
• Max iter (from, to) - maximum number of network teaching repetitions (the same data is used for learning many times in internal processes of the neural network). The slower it is. Jump every 100. The maximum value is 10, the default is 1.
• Split data to train&test - for example, the entered value of 0.8 means that the input data for the network will be divided in the ratio 0.8 to learning, 0.2 for the tests of the network learned.
• Data limit - limits the amount of data downloaded from the source. It speeds up the processing, but reduces its quality.
• Max probes - limits the number of samples taken to learn the network. Samples are already aggregated according to the selected “Time frame” parameter. It speed up teaching but reduces its quality.
• Build - a button to start teaching the network. The button contains the number of required teaching curses. You should be careful and avoid one-time learning for more than 1000 courses. It is better to divide them into several smaller ones. One pass after a full data load take about 1-3 minutes on a 4 core 2.4.GHz server. The module has implemented the best practices related to the number of neurons in individual hidden layers. The values suggested by the system are optimal from the point of view of these practices, but the user can decide on these values himself.

Under the parameters for learning the network there is an area in which teaching results will appear.

After pressing the “Refresh” button, the list of the resulting models will be refreshed.

Autorefresh - selecting the field automatically refreshes the list of learning results every 10s.

The following information will be available in the table on the left:

• Internal name - the model name given by the system, including the user - specified prefix
• Overall efficiency - the network adjustment indicator - allow to see at a glance whether it is worth dealing with the model. The grater the value, the better.

After clicking on the table row, detailed data collected during the learning of the given model will be displayed. This data will be visible in the box on the right.

The selected model can be saved under its own name using the “Save algorithm” button. This saved algorithm will be available in the “Choose AI Rule” list when creating the rule (see Create AI Rule).

AI Learn Tasks

The “AI Learn Task” tab shows the list of processes initiated teaching the ANN network with the possibility of managing processes.

Each user can see only the process they run. The user in the role of Intelligence sees all running processes.

Description of controls:

• Algorithm prefix - this is the value set by the user on the AI Learn screen in the Prefix name field
• Progress - here is the number of algorithms generated / the number of all to be generated
• Processing time - duration of algorithm generation in seconds (or maybe minutes or hours)
• Actions:
  • Cancel - deletes the algorithm generation task (user require confirmation of operation)
  • Pause / Release - pause / resume algorithm generation process.

AI Learn tab contain the Show in the preview mode of the ANN hyperparameters After completing the learning activity or after the user has interrupted it, the “Delete” button appears in “Action” field. This button allows you to permanently delete the learning results of a specific network.

Scenarios of using algorithms implemented in the Intelligence module

Teaching MLP networks and choosing the algorithm to use:

1. Go to the AI Learn tab,
2. We introduce the network teaching parameters,
3. Enter your own prefix for the names of the algorithms you have learned,
4. Press Build.
5. We observe the learned networks on the list (we can also stop the observation at any moment and go to other functions of the system. We will return to the learning results by going to the AI Learn Tasks tab and clicking the show action),
6. We choose the best model from our point of view and save it under our own name,
7. From this moment the algorithm is visible in the Create AI Rule tab.

Starting the MLP network algorithm:

1. Go to the Create AI Rule tab and create rules,
2. Select the previously saved model of the learned network,
3. Specify parameters visible on the screen (specific to MLP),
4. Press the Run button.

Starting regression algorithm:

1. Go to the Create AI Rule tab and create rules,
2. We choose AI Rule, e.g. Simple Moving Average, Linear Regression or Random Forest Regression, etc.,
3. Enter your own rule name (specific to regression),
4. Set the parameters of the rule ( specific to regression),
5. Press the Run button.

Management of available rules:

1. Go to the AI Rules List tab,
2. A list of AI rules available for our role is displayed,
3. We can perform the actions available on the right for each rule.# Results of algorithms #

The results of the “AI algorithms” are saved to the index „intelligence” specially created for this purpose. The index with the prediction result. These following fields are available in the index (where xxx is the name of the attribute being analyzed):

• xxx_pre - estimate value
• xxx_cur - current value at the moment of estimation
• method_name - name of the algorithm used
• rmse - avarage square error for the analysis in which _cur values were available. The smaller the value, the better.
• rmse_normalized - mean square error for the analysis in which _cur values were available, normalized with _pre values. The smaller the value, the better.
• overall_efficiency - efficiency of the model. The greater the value, the better. A value less than 0 may indicate too little data to correctly calculate the indicator
• linear_function_a - directional coefficient of the linear function y = ax + b. Only for the Trend and Linear Regression Trend algorithm
• linear_function_b - the intersection of the line with the Y axis for the linear function y = ax + b. Only for the Trend and Linear Regression Trend algorithm.

Visualization and signals related to the results of data analysis should be created from this index. The index should be available to users of the Intelligence module.

Scheduler Module

OP5 Log Analytics has a built-in task schedule. In this module, we can define a command or a list of commands whose execution we instruct the application in the form of tasks. We can determine the time and frequency of tasks. Tasks can contain a simple syntax, but they can also be associated with modules, e.g. with Intelligence module.

To go to the Scheduler window, select the tile icon from the main menu bar and then go to the „Scheduler” icon (To go back, go to the „Search” icon)

The page with three tabs will be displayed: Creating new tasks in the „Create Scheduler Job”, managing tasks in the „Job List” and checking the status of tasks in „Jobs Status”

In the window for creating new tasks we have a form consisting of fields:

• Name - in which we enter the name of the task
• Cron Pattern - a field in which in cron notation we define the time and frequency of the task
• Command - we give the syntax of the command that will be executed in this task. These can be simple system commands, but also complex commands related to the Intelligence module. In the task management window, we can activate /deactivate, delete and update the task by clicking on the selected icon for a given task

In the task status windows you can check the current status of the task: if it activated, when it started and when it ended, how long it took. This window is not editable and indicates historical data.

Permission

Permission have been implemented in the following way:

• Only the user in the admin role can create / update rules.
• When creating rules, the roles that will be able to enables / disengage / view the rules will be indicated.

We assume that the Learn process works as an administrator.

We assume that the visibility of Search in AI Learn is preceded by receiving the search permission in the module object permission.

The role of “Intelligence” launches the appropriate tabs.

An ordinary user only sees his models. The administrator sees all models.

Register new algorithm

For register new algorithm:

• Login to the OP5 Log Analytics
• Select Intelligence
• Select Algorithm
• Fill Create algorithm form and press Submit button

Form fields:

| Field   | Description                                                                                                      |
|---------|------------------------------------------------------------------------------------------------------------------|
| Code    | Short name for algorithm                                                                                         |
| Name    | Algorithm name                                                                                                   |
| Command | Command to execute. The command must be in the directory pointed to by the parameter elastscheduler.commandpath. |

OP5 Log Analytics execute command:

<command> <config> <error file> <out file>

Where:

• command - Command from command filed of Create algorithm form.
• config - Full path of json config file. The name of file is id of process status document in index .intelligence_rules
• error file - Unique name for error file. Not used by predefined algorithms.
• out file - Unique name for output file. Not used by predefined algorithms.

Config file:

Json document:

| Field                  | Value                                                                               | Screen field (description)                                           |
|------------------------|-------------------------------------------------------------------------------------|----------------------------------------------------------------------|
| algorithm_type         | GMA, GMAL, LRS, LRST, RFRS, SMAL, SMA, TL                                           | Algorithm. For customs method field Code from Create algorithm form. |
| model_name             | Not empty string.                                                                   | AI Rule Name.                                                        |
| search                 | Search id.                                                                          | Choose search.                                                       |
| label_field.field      |                                                                                     | Feature to analyse.                                                  |
| max_probes             | Integer value                                                                       | Max probes                                                           |
| time_frame             | 1 minute, 5 minutes, 15 minutes, 30 minutes, 1 hour, 1 day, 1 week, 30 day, 365 day | Time frame                                                           |
| value_type             | min, max, avg, count                                                                | Value type                                                           |
| max_predictions        | Integer value                                                                       | Max predictions                                                      |
| threshold              | Integer value                                                                       | Threshold                                                            |
| automatic_cron         | Cron format string                                                                  | Automatic cycle                                                      |
| automatic_enable       | true/false                                                                          | Enable                                                               |
| automatic              | true/false                                                                          | Automatic                                                            |
| start_date             | YYYY-MM-DD HH:mm or now                                                             | Start date                                                           |
| multiply_by_values     | Array of string values                                                              | Multiply by values                                                   |
| multiply_by_field      | None or full field name eg.: system.cpu                                             | Multiply by field                                                    |
| selectedroles          | Array of roles name                                                                 | Role                                                                 |
| last_execute_timestamp |                                                                                     | Last execute                                                         |

| Not screen fields     |                                     |
|-----------------------|-------------------------------------|
| preparation_date      | Document preparation date.          |
| machine_state_uid     | AI rule machine state uid.          |
| path_to_logs          | Path to ai machine logs.            |
| path_to_machine_state | Path to ai machine state files.     |
| searchSourceJSON      | Query string.                       |
| processing_time       | Process operation time.             |
| last_execute_mili     | Last executed time in milliseconds. |
| pid                   | Process pid if ai rule is running.  |
| exit_code             | Last executed process exit code.    |

The command must update the process status document in the system during operation. It is elastic partial document update.

| Process status         | Field (POST body)          | Description                                    |
|------------------------|----------------------------|------------------------------------------------|
| START                  | doc.pid                    | System process id                              |
|                        | doc.last_execute_timestamp | Current timestamp. yyyy-MM-dd HH:mm            |
|                        | doc.last_execute_mili      | Current timestamp in millisecunds.             |
| END PROCESS WITH ERROR | doc.error_description      | Error description.                             |
|                        | doc.error_message          | Error message.                                 |
|                        | doc.exit_code              | System process exit code.                      |
|                        | doc.pid                    | Value “0”.                                       |
|                        | doc.processing_time        | Time of execute process in seconds.            |
| END PROCESS OK         | doc.pid                    | Value “0”.                                       |
|                        | doc.exit_code              | System process exit code. Value “0” for success. |
|                        | doc.processing_time        | Time of execute process in seconds.            |

The command must insert data for prediction chart.

| Field             | Value             | Description                                            |
|-------------------|-------------------|--------------------------------------------------------|
| model_name        | Not empty string. | AI Rule Name.                                          |
| preparationUID    | Not empty string. | Unique prediction id                                   |
| machine_state_uid | Not empty string. | AI rule machine state uid.                             |
| model_uid         | Not empty string. | Model uid from config file                             |
| method_name       | Not empty string. | User friendly algorithm name.                          |
| <field>           | Json              | Field calculated. For example: system.cpu.idle.pct_pre |

Document sample:

		{
		  "_index": "intelligence",
		  "_type": "doc",
		  "_id": "emca_TL_20190304_080802_20190531193000",
		  "_version": 2,
		  "_score": null,
		  "_source": {
		    "machine_state_uid": "emca_TL_20190304_080802",
		    "overall_efficiency": 0,
		    "processing_time": 0,
		    "rmse_normalized": 0,
		    "predictionUID": "emca_TL_20190304_080802_20190531193000",
		    "linear_function_b": 0,
		    "@timestamp": "2019-05-31T19:30:00.000+0200",
		    "linear_function_a": 0.006787878787878788,
		    "system": {
		      "cpu": {
		        "idle": {
		          "pct_pre": 0.8213333333333334
		        }
		      }
		    },
		    "model_name": "emca",
		    "method_name": "Trend",
		    "model_uid": "emca_TL_20190304_080802",
		    "rmse": 0,
		    "start_date": "2019-03-04T19:30:01.279+0100"
		  },
		  "fields": {
		    "@timestamp": [
		      "2019-05-31T17:30:00.000Z"
		    ]
		  },
		  "sort": [
		    1559323800000
		  ]
		}

Verification steps and logs

Verification of Elasticsearch service

To verify of Elasticsearch service you can use following command:

• Control of the Elastisearch system service via systemd:

    # sysetmctl status elasticsearch
  

  output:

      ● elasticsearch.service - Elasticsearch
       Loaded: loaded (/usr/lib/systemd/system/elasticsearch.service; disabled; vendor preset: disabled)
       Active: active (running) since Mon 2018-09-10 13:11:40 CEST; 22h ago
     Docs: http://www.elastic.co
     Main PID: 1829 (java)
       CGroup: /system.slice/elasticsearch.service
       └─1829 /bin/java -Xms4g -Xmx4g -XX:+UseConcMarkSweepGC -XX:CMSInitiatingOccupancyFraction=75 -XX:+UseCMSInitiatingOccupancyOnly -XX:+AlwaysPreTouch -Xss1m ...
  

• Control of Elasticsearch instance via tcp port:

     # curl -XGET '127.0.0.1:9200/'
  

  output:

    	{
    	  "name" : "dY3RuYs",
    	  "cluster_name" : "elasticsearch",
    	  "cluster_uuid" : "EHZGAnJkStqlgRImqwzYQQ",
    	  "version" : {
    	    "number" : "6.2.3",
    	    "build_hash" : "c59ff00",
    	    "build_date" : "2018-03-13T10:06:29.741383Z",
    	    "build_snapshot" : false,
    	    "lucene_version" : "7.2.1",
    	    "minimum_wire_compatibility_version" : "5.6.0",
    	    "minimum_index_compatibility_version" : "5.0.0"
    	  },
    	  "tagline" : "You Know, for Search"
    	}
  

• Control of Elasticsearch instance via log file:

    # tail -f /var/log/elasticsearch/elasticsearch.log
  

• other control commands via curl application:

    curl -xGET "http://localhost:9200/_cat/health?v"
    curl -XGET "http://localhost:9200/_cat/nodes?v"
    curl -XGET "http://localhost:9200/_cat/indicates"
  

Verification of Logstash service

To verify of Logstash service you can use following command:

• control Logstash service via systemd:

    # systemctl status logstash
  

output:

	logstash.service - logstash
	   Loaded: loaded (/etc/systemd/system/logstash.service; enabled; vendor preset: disabled)
	   Active: active (running) since Wed 2017-07-12 10:30:55 CEST; 1 months 23 days ago
	 Main PID: 87818 (java)
	   CGroup: /system.slice/logstash.service
	          └─87818 /usr/bin/java -XX:+UseParNewGC -XX:+UseConcMarkSweepGC

• control Logstash service via port tcp:

    # curl -XGET '127.0.0.1:9600'
  

output:

	{
	   "host": "skywalker",
	   "version": "4.5.3",
	   "http_address": "127.0.0.1:9600"
	}

• control Logstash service via log file:

    # tail -f /var/log/logstash/logstash-plain.log
  

Debuging

• dynamically update logging levels through the logging API (service restart not needed):

     curl -XPUT 'localhost:9600/_node/logging?pretty' -H 'Content-Type: application/json' -d'
     {
         "logger.logstash.outputs.elasticsearch" : "DEBUG"
     }
     '
  

• permanent change of logging level (service need to be restarted):

  • edit file /etc/logstash/logstash.yml and set the following parameter:

       *log.level: debug*
    

  • restart logstash service:

     	*systemctl restart logstash*
    

• checking correct syntax of configuration files:

    */usr/share/logstash/bin/logstash -tf /etc/logstash/conf.d*
  

• get information about load of the Logstash:

    *# curl -XGET '127.0.0.1:9600/_node/jvm?pretty=true'*
  

output:

 {                                                          
  "host" : "logserver-test",                               
  "version" : "5.6.2",                                     
  "http_address" : "0.0.0.0:9600",                         
  "id" : "5a440edc-1298-4205-a524-68d0d212cd55",           
  "name" : "logserver-test",                               
  "jvm" : {                                                
    "pid" : 14705,                                         
    "version" : "1.8.0_161",                               
    "vm_version" : "1.8.0_161",                            
    "vm_vendor" : "Oracle Corporation",                    
    "vm_name" : "Java HotSpot(TM) 64-Bit Server VM",       
    "start_time_in_millis" : 1536146549243,                
    "mem" : {                                              
      "heap_init_in_bytes" : 268435456,                    
      "heap_max_in_bytes" : 1056309248,                    
      "non_heap_init_in_bytes" : 2555904,                  
      "non_heap_max_in_bytes" : 0                          
    },                                                     
    "gc_collectors" : [ "ParNew", "ConcurrentMarkSweep" ]  
  }                                                        # Verificatoin of OP5 Log Analytics GUI service #

To verify of OP5 Log Analytics GUI service you can use following command:

• control the OP5 Log Analytics GUI service via systemd:

    # systemctl status kibana
  

output:

	● kibana.service - Kibana                                                                                                         
	   Loaded: loaded (/etc/systemd/system/kibana.service; disabled; vendor preset: disabled)                                         
	   Active: active (running) since Mon 2018-09-10 13:13:19 CEST; 23h ago                                                           
	 Main PID: 1330 (node)                                                                                                            
	   CGroup: /system.slice/kibana.service                                                                                           
	           └─1330 /usr/share/kibana/bin/../node/bin/node --no-warnings /usr/share/kibana/bin/../src/cli -c /etc/kibana/kibana.yml 

• control the OP5 Log Analytics GUI via port tcp/http:

    # curl -XGET '127.0.0.1:5601/'
  

output:

	  <script>var hashRoute = '/app/kibana';
	  var defaultRoute = '/app/kibana';
	  var hash = window.location.hash;
	  if (hash.length) {
	    window.location = hashRoute + hash;
	  } else {
	    window.location = defaultRoute;
	  }</script>

• Control the OP5 Log Analytics GUI via log file:

    # tail -f /var/log/messages
  

Building a cluster

Node roles

Every instance of Elasticsearch server is called a node. A collection of connected nodes is called a cluster. All nodes know about all the other nodes in the cluster and can forward client requests to the appropriate node.

Besides that, each node serves one or more purpose:

• Master-eligible node - A node that has node.master set to true (default), which makes it eligible to be elected as the master node, which controls the cluster
• Data node - A node that has node.data set to true (default). Data nodes hold data and perform data related operations such as CRUD, search, and aggregations
• Client node - A client node has both node.master and node.data set to false. It can neither hold data nor become the master node. It behaves as a “smart router” and is used to forward cluster-level requests to the master node and data-related requests (such as search) to the appropriate data nodes
• Tribe node - A tribe node, configured via the tribe.* settings, is a special type of client node that can connect to multiple clusters and perform search and other operations across all connected clusters.

Naming convention

Elasticsearch require little configuration before before going into work.

The following settings must be considered before going to production:

• path.data and path.logs - default locations of these files are: /var/lib/elasticsearchand /var/log/elasticsearch.
• cluster.name - A node can only join a cluster when it shares its cluster.name with all the other nodes in the cluster. The default name is “elasticsearch”, but you should change it to an appropriate name which describes the purpose of the cluster. You can do this in /etc/elasticsearch/elasticsearch.yml file.
• node.name - By default, Elasticsearch will use the first seven characters of the randomly generated UUID as the node id. Node id is persisted and does not change when a node restarts. It is worth configuring a more human readable name: node.name: prod-data-2 in file /etc/elstaicsearch/elasticsearch.yml
• network.host - parametr specifying network interfaces to which Elasticsearch can bind. Default is network.host: ["_local_","_site_"].
• discovery - Elasticsearch uses a custom discovery implementation called “Zen Discovery”. There are two important settings:
  • discovery.zen.ping.unicast.hosts - specify list of other nodes in the cluster that are likely to be live and contactable;
  • discovery.zen.minimum_master_nodes - to prevent data loss, you can configure this setting so that each master-eligible node knows the minimum number of master-eligible nodes that must be visible in order to form a cluster.
• heap size - By default, Elasticsearch tells the JVM to use a heap with a minimum (Xms) and maximum (Xmx) size of 1 GB. When moving to production, it is important to configure heap size to ensure that Elasticsearch has enough heap available

Config files

To configure the Elasticsearch cluster you must specify some parameters in the following configuration files on every node that will be connected to the cluster:

• /etc/elsticsearch/elasticserach.yml:
  • cluster.name:name_of_the_cluster - same for every node;
  • node.name:name_of_the_node - uniq for every node;
  • node.master:true_or_false
  • node.data:true_or_false
  • network.host:["_local_","_site_"]
  • discovery.zen.ping.multicast.enabled
  • discovery.zen.ping.unicast.hosts
• /etc/elsticsearch/log4j2.properties:
  • logger: action: DEBUG - for easier debugging.

Example setup

Example of the Elasticsearch cluster configuration:

• file /etc/elasticsearch/elasticsearch.yml:

    cluster.name: tm-lab
    node.name: "elk01"
    node.master: true
    node.data: true
    network.host: 127.0.0.1,10.0.0.4
    http.port: 9200
    discovery.zen.ping.multicast.enabled: false
    discovery.zen.ping.unicast.hosts: ["10.0.0.4:9300","10.0.0.5:9300","10.0.0.6:9300"]
  

• to start the Elasticsearch cluster execute command:

    # systemctl restart elasticsearch
  

• to check status of the Elstaicsearch cluster execute command:

  • check of the Elasticsearch cluster nodes status via tcp port:

      # curl -XGET '127.0.0.1:9200/_cat/nodes?v'
    
      host         	  ip           heap.percent ram.percent load node.role master name
      10.0.0.4 	 10.0.0.4     18           91 		   0.00 -         -      elk01
      10.0.0.5 	 10.0.0.5     66           91 		   0.00 d        *      elk02
      10.0.0.6 	 10.0.0.6     43           86         	   0.65 d        m     elk03
      10.0.0.7 	 10.0.0.7     45           77         	   0.26 d        m     elk04
    

  • check status of the Elasticsearch cluster via log file:

      # tail -f /var/log/elasticsearch/tm-lab.log (cluster.name)
    

Adding a new node to existing cluster

Install the new Energy instance. The description of the installation can be found in the chapter “First configuration steps”

Change the following parameters in the configuration file:

• cluster.name:name_of_the_cluster same for every node;
• node.name:name_of_the_node uniq for every node;
• node.master:true_or_false
• node.data:true_or_false
• discovery.zen.ping.unicast.hosts:[“10.0.0.4:9300”,”10.0.0.5:9300”,”10.0.0.6:9300”] - IP addresses and instances of nodes in the cluster.

If you add a node with the role data, delete the contents of the path.data directory, by default in /var/lib/elasticsearch

Restart the Elasticsearch instance of the new node:

systemctl restart elasticsearch

Integration with AD

You can configure the OP5 Log Analytics to communicate with Active Directory to authenticate users. To integrate with Active Directory, you configure an Active Directory realm and assign Active Directory users and groups to the OP5 Log Analytics roles in the role mapping file.

To protect passwords, communications between the OP5 Log Analytics and the LDAP server should be encrypted using SSL/TLS. Clients and nodes that connect via SSL/TLS to the LDAP server need to have the LDAP server’s certificate or the server’s root CA certificate installed in their keystore or truststore.

AD configuration

The AD configuration should be done in the /etc/elasticsearch/properties.yml file.

Below is a list of settings to be made in the properties.yml file (the commented section in the file in order for the AD settings to start working, this fragment should be uncommented):

|**Direcitve**                          		| **Description**               							|
| ------------------------------------------------------|---------------------------------------------------------------------------------------|
| # LDAP                                		|                              								|
| #ldaps:                               		|                               							|
| # - name: \"example.com\"             		|# domain that is configured    							|
| # host: \"127.0.0.1,127.0.0.2\"       		|# list of server for this domain							|
| # port: 389                           		|# optional, default 389 for unencrypted session or 636 for encrypted sessions		|
|# ssl\_enabled: false                  		|# optional, default true       							|
|# ssl\_trust\_all\_certs: true         		|# optional, default false      							|
|# ssl.keystore.file: \"path\"          		|# path to the truststore store 							|
|# ssl.keystore.password: \"path\"      		|# password to the trusted certificate store  						|
|# bind\_dn: [[admin\@example.com]      		|# account name administrator   							|
|# bind\_password: \"password\"         		|# password for the administrator account 						|
|# search\_user\_base\_DN: \"OU=lab,DC=example,DC=com\" |# search for the DN user tree database 						|
|# user\_id\_attribute: \"uid           		|# search for a user attribute optional, by default \"uid\"            			|
|# search\_groups\_base\_DN:\"OU=lab,DC=example,DC=com\"|# group database search. This is a catalog main, after which the groups will be sought.|
|# unique\_member\_attribute: \"uniqueMember\" 		|# optional, default\"uniqueMember\"							|
|# connection\_pool\_size: 10                  		|# optional, default 30									|
|# connection\_timeout\_in\_sec: 10                  	|# optional, default 1									|
|# request\_timeout\_in\_sec: 10                     	|# optional, default 1									|
|# cache\_ttl\_in\_sec: 60                           	|# optional, default 0 - cache disabled							|

If we want to configure multiple domains, then in this configuration file we copy the # LDAP section below and configure it for the next domain.

Below is an example of how an entry for 2 domains should look like. (It is important to take the interpreter to read these values ​​correctly).

ldaps:
        - name: "example1.com"
          host: "127.0.0.1,127.0.0.2"
          port: 389 # optional, default 389
          ssl_enabled: false # optional, default true
          ssl_trust_all_certs: true # optional, default false
          bind_dn: "admin@example1.com"
          bind_password: "password" # generate encrypted password with /usr/share/elasticsearch/pass-encrypter/pass-encrypter.sh
          search_user_base_DN: "OU=lab,DC=example1,DC=com"
          user_id_attribute: "uid" # optional, default "uid"
          search_groups_base_DN: "OU=lab,DC=example1,DC=com"
          unique_member_attribute: "uniqueMember" # optional, default "uniqueMember"
          connection_pool_size: 10 # optional, default 30
          connection_timeout_in_sec: 10 # optional, default 1
          request_timeout_in_sec: 10 # optional, default 1
          cache_ttl_in_sec: 60 # optional, default 0 - cache disabled
          service_principal_name: "esauth@example1.com" # optional, for sso
          service_principal_name_password : "password" # optional, for sso
        - name: "example2.com" #DOMAIN 2
          host: "127.0.0.1,127.0.0.2"
          port: 389 # optional, default 389
          ssl_enabled: false # optional, default true
          ssl_trust_all_certs: true # optional, default false
          bind_dn: "admin@example2.com"
          bind_password: "password" # generate encrypted password with /usr/share/elasticsearch/pass-encrypter/pass-encrypter.sh
          search_user_base_DN: "OU=lab,DC=example2,DC=com"
          user_id_attribute: "uid" # optional, default "uid"
          search_groups_base_DN: "OU=lab,DC=example2,DC=com"
          unique_member_attribute: "uniqueMember" # optional, default "uniqueMember"
          connection_pool_size: 10 # optional, default 30
          connection_timeout_in_sec: 10 # optional, default 1
          request_timeout_in_sec: 10 # optional, default 1
          cache_ttl_in_sec: 60 # optional, default 0 - cache disabled
          service_principal_name: "esauth@example2.com" # optional, for sso
          service_principal_name_password : "password" # optional, for ssl

After completing the LDAP section entry in the properties.yml file, save the changes and restart the service with the command:

# systemctl restart elasticsearch

Configure SSL suport for AD authentication

Open the certificate manager on the AD server.

Select the certificate and open it

Select the option of copying to a file in the Details tab

Click the Next button

Keep the setting as shown below and click Next

Keep the setting as shown below and click Next.

Give the name a certificate

After the certificate is exported, this certificate should be imported into a trusted certificate file that will be used by the Elasticsearch plugin.

To import a certificate into a trusted certificate file, a tool called „keytool.exe” is located in the JDK installation directory.

Use the following command to import a certificate file:

keytool -import -alias adding_certificate_keystore -file certificate.cer -keystore certificatestore

The values for RED should be changed accordingly.

By doing this, he will ask you to set a password for the trusted certificate store. Remember this password, because it must be set in the configuration of the Elasticsearch plugin. The following settings must be set in the properties.yml configuration for SSL:

ssl.keystore.file: "<path to the trust certificate store>"
ssl.keystore.password: "< password to the trust certificate store>"

Role mapping

In the /etc/elasticsearch/properties.yml configuration file you can find a section for configuring role mapping:

# LDAP ROLE MAPPING FILE`
# rolemapping.file.path: /etc/elasticsearch/role-mappings.yml

This variable points to the file /etc/elasticsearch/role-mappings.yml Below is the sample content for this file:

admin:

"CN=Admins,OU=lab,DC=dev,DC=it,DC=example,DC=com"

bank:

"CN=security,OU=lab,DC=dev,DC=it,DC=example,DC=com"

How to the mapping mechanism works ? An AD user log in to OP5 Log Analytics. In the application there is a admin role, which through the file role-mapping .yml binds to the name of the admin role to which the Admins container from AD is assigned. It is enough for the user from the AD account to log in to the application with the privileges that are assigned to admin role in the OP5 Log Analytics. At the same time, if it is the first login in the OP5 Log Analytics, an account is created with an entry that informs the application administrator that is was created by logging in with AD.

Similar, the mechanism will work if we have a role with an arbitrary name created in OP5 Logistics and connected to the name of the role-mappings.yml and existing in AD any container.

Below a screenshot of the console on which are marked accounts that were created by uesrs logging in from AD

If you map roles with from several domains, for example dev.examloe1.com, dev.example2.com then in User List we will see which user from which domain with which role logged in OP5 Log Analytics.

Password encryption

For security reason you can provide the encrypted password for Active Directory integration. To do this use pass-encrypter.sh script that is located in the Utils directory in installation folder.

1. Installation of pass-encrypter

    cp -pr /instalation_folder/elasticsearch/pass-encrypter /usr/share/elasticsearch/
   

2. Use pass-encrypter

    # /usr/share/elasticsearch/pass-encrypter/pass-encrypter.sh
    Enter the string for encryption :
    new_password
    Encrypted string : MTU1MTEwMDcxMzQzMg==1GEG8KUOgyJko0PuT2C4uw==
   

Integration with Radius

To use the Radius protocol, install the latest available version of OP5 Log Analytics.

Configuration

The default configuration file is located at /etc/elasticsearch/properties.yml:

	# Radius opts
	#radius.host: "10.4.3.184"
	#radius.secret: "querty1q2ww2q1"
	#radius.port: 1812

Use appropriate secret based on config file in Radius server. The secret is configured on clients.conf in Radius server.

In this case, since the plugin will try to do Radius auth then client IP address should be the IP address where the Elasticsearch is deployed.

Every user by default at present get the admin role.

Configuring Single Sign On (SSO)

In order to configure SSO, the system should be accessible by domain name URL, not IP address nor localhost.

Ok :https://loggui.com:5601/login. Wrong : https://localhost:5601/login, https://10.0.10.120:5601/login

In order to enable SSO on your system follow below steps. The configuration is made for AD: dev.example.com, GUI URL: loggui.com

Configuration steps

Create an User Account for Elasticsearch auth plugin

In this step, a Kerberos Principal representing Elasticsearch auth plugin is created on the Active Directory. The principal name would be name@DEV.EXAMPLE.COM, while the DEV.EXAMPLE.COM is the administrative name of the realm. In our case, the principal name will be esauth@DEV.EXAMPLE.COM.

Create User in AD. Set “Password never expires” and “Other encryption options” as shown below:

Define Service Principal Name (SPN) and Create a Keytab file for it

Use the following command to create the keytab file and SPN:

C:> ktpass -out c:\Users\Administrator\esauth.keytab -princ HTTP/loggui.com@DEV.EXAMPLE.COM -mapUser esauth -mapOp set -pass ‘Sprint$123’ -crypto ALL -pType KRB5_NT_PRINCIPAL

Values highlighted in bold should be adjusted for your system. The esauth.keytab file should be placed on your elasticsearch node - preferably /etc/elasticsearch/ with read permissions for elasticsearch user: chmod 640 /etc/elasticsearch/esauth.keytab chown elasticsearch: /etc/elasticsearch/esauth.keytab

Create a file named krb5Login.conf:

com.sun.security.jgss.initiate{
    com.sun.security.auth.module.Krb5LoginModule required
    principal="esauth@DEV.EXAMPLE.COM" useKeyTab=true
    keyTab=/etc/elasticsearch/esauth.keytab storeKey=true debug=true;
    };
com.sun.security.jgss.krb5.accept {
    com.sun.security.auth.module.Krb5LoginModule required
    principal="esauth@DEV.EXAMPLE.COM" useKeyTab=true
    keyTab=/etc/elasticsearch/esauth.keytab storeKey=true debug=true;
    };

Principal user and keyTab location should be changed as per the values created in the step 2. Make sure the domain is in UPPERCASE as shown above. The krb5Login.conf file should be placed on your elasticsearch node, for instance /etc/elasticsearch/ with read permissions for elasticsearch user:

sudo chmod 640 /etc/elasticsearch/krb5Login.conf
sudo chown elasticsearch: /etc/elasticsearch/krb5Login.conf

Append the following JVM arguments (on Elasticsearch node in /etc/sysconfig/elasticsearch):

-Dsun.security.krb5.debug=true -Djava.security.krb5.realm=DEV.EXAMPLE.COM -Djava.security.krb5.kdc=AD_HOST_IP_ADDRESS -Djava.security.auth.login.config=/etc/elasticsearch/krb5Login.conf -Djavax.security.auth.useSubjectCredsOnly=false

Change the appropriate values in the bold. This JVM arguments has to be set for Elasticsearch server.

Add the following additional (sso.domain, service_principal_name, service_principal_name_password) settings for ldap in elasticsearch.yml or properties.yml file wherever the ldap settings are configured:

sso.domain: "dev.example.com"
ldaps:
- name: "dev.example.com"
    host: "IP_address"
    port: 389                                                 # optional, default 389
    ssl_enabled: false                                        # optional, default    true
    ssl_trust_all_certs: false                                 # optional, default false
    bind_dn: "Administrator@dev.example.com"                     # optional, skip for anonymous bind
    bind_password: "administrator_password"                                 # optional, skip for anonymous bind
    search_user_base_DN: "OU=lab,DC=dev,DC=it,DC=example,DC=com"
    user_id_attribute: "uid"                                  # optional, default "uid"
    search_groups_base_DN: "OU=lab,DC=dev,DC=it,DC=example,DC=com"
    unique_member_attribute: "uniqueMember"                   # optional, default "uniqueMember"
    service_principal_name: "esauth@DEV.EXAMPLE.COM"
    service_principal_name_password : "Sprint$123"

Note: At this moment, SSO works for only single domain. So you have to mention for what domain SSO should work in the above property sso.domain

To apply the changes restart Elasticsearch service

sudo systemctl restart elasticsearch.service

Enable SSO feature in kibana.yml file:

kibana.sso_enabled: true After that Kibana has to be restarted: sudo systemctl restart kibana.service

Client (Browser) Configuration##

Internet Explorer configuration

1. Goto Internet Options from Tools menu and click on Security Tab:

1. Select Local intranet, click on Site -> Advanced -> Add the url:

After adding the site click close.

1. Click on custom level and select the option as shown below:

Chrome configuration

For Chrome, the settings are taken from IE browser.

Firefox configuration

Update the following config:

Configure email delivery

Configure email delivery for sending PDF reports in Scheduler.

The default e-mail client that installs with the Linux CentOS system, which is used by OP5 Log Analytics to send reports (Section 5.3 of the Reports chapter), is postfix.# Configuration file for postfix mail client #

The postfix configuration directory for CentOS is /etc/postfix. It contains files:

main.cf - the main configuration file for the program specifying the basics parameters

Some of its directives:

 |**Directive**            |       **Description**                                                                                    |
 | ------------------------| ---------------------------------------------------------------------------------------------------------|
 |queue\_directory         |       The postfix queue location.                                                                         
 |command\_directory       |      The location of Postfix commands.
 |daemon\_directory        |       Location of Postfix daemons.
 |mail\_owner              |       The owner of Postfix domain name of the server
 |myhostname               |       The fully qualified domain name of the server.
 |mydomain                 |       Server domain
 |myorigin                 |       Host or domain to be displayed as origin on email leaving the server.
 |inet\_interfaces         |       Network interface to be used for incoming email.
 |mydestination            |       Domains from which the server accepts mail.
 |mynetworks               |       The IP address of trusted networks.
 |relayhost                |       Host or other mail server through which mail will be sent. This server will act as an outbound gateway.
 |alias\_maps              |       Database of asliases used by the local delivery agent.
 |alias\_database          |       Alias database generated by the new aliases command.
 |mail\_spool\_directory   |       The location where user boxes will be stored.

master.cf - defines the configuration settings for the master daemon and the way it should work with other agents to deliver mail. For each service installed in the master.cf file there are seven columns that define how the service should be used.

 |Column           |     Description
 |---------------- | --------------------------------------------------------------------------------------------
 |service          |    The name of the service
 |type             |    The transport mechanism to be user.
 |private          |    Is the service only for user by Postfix.
 |unpriv           |    Can the service be run by ordinary users
 |chroot           |    Whether the service is to change the main directory (chroot) for the mail. Queue.
 |wakeup           |    Wake up interval for the service.
 |maxproc          |    The maximum number of processes on which the service can be forked (to divide in branches)
 |command + args   |   A command associated with the service plus any argument

access - can be used to control access based on e-mail address, host address, domain or network address.

Examples of entries in the file

 |Description                                     | Example
 |------------------------------------------------|--------------------
 |To allow access for specific IP address:        | 192.168.122.20 OK
 |To allow access for a specific domain:          | example.com OK
 |To deny access from the 192.168.3.0/24 network: | 192.168.3 REJECT

After making changes to the access file, you must convert its contents to the access.db database with the postmap command:

 	# postmap /etc/postfix/access
 	# ll /etc/postfix/access*
 
 	-rw-r\--r\--. 1 root root 20876 Jan 26 2014 /etc/postfix/access
 	-rw-r\--r\--. 1 root root 12288 Feb 12 07:47 /etc/postfix/access.db

canonical - mapping incoming e-mails to local users.

Examples of entries in the file:

To forward emails to user1 to the [[user1@yahoo.com] mailbox:

	user1 user1\@yahoo.com

To forward all emails for example.org to another example.com domain:

	@example.org @example.com

After making changes to the canonical file, you must convert its contents to the canonical.db database with the postmap command:

# postmap /etc/postfix/canonical
# ll /etc/postfix/canonical*

-rw-r\--r\--. 1 root root 11681 2014-06-10 /etc/postfix/canonical
-rw-r\--r\--. 1 root root 12288 07-31 20:56 /etc/postfix/canonical.db

generic - mapping of outgoing e-mails to local users. The syntax is the same as a canonical file. After you make change to this file, you must also run the postmap command.

# postmap /etc/postfix/generic
# ll /etc/postfix/generic*

-rw-r\--r\--. 1 root root 9904 2014-06-10 /etc/postfix/generic
-rw-r\--r\--. 1 root root 12288 07-31 21:15 /etc/postfix/generic.db

reloceted – information about users who have been transferred. The syntax of the file is the same as canonical and generic files.

Assuming tha user1 was moved from example.com to example.net, you can forward all emails received on the old address to the new address:

Example of an entry in the file:

user1@example.com user1@example.net

After you make change to this file, you must also run the postmap command.

# postmap /etc/postfix/relocated
# ll /etc/postfix/relocated*

-rw-r\--r\--. 1 root root 6816 2014-06-10 /etc/postfix/relocated
-rw-r\--r\--. 1 root root 12288 07-31 21:26 /etc/postfix/relocated.d

transport – mapping between e-mail addresses and server through which these e-mails are to be sent (next hops) int the transport format: nexthop.

Example of an entry in the file:

user1@example.com smtp:host1.example.com

After you make changes to this file, you must also run the postmap command.

# postmap /etc/postfix/transport
[root@server1 postfix]# ll /etc/postfix/transport*

-rw-r\--r\--. 1 root root 12549 2014-06-10 /etc/postfix/transport
-rw-r\--r\--. 1 root root 12288 07-31 21:32 /etc/postfix/transport.db

virtual - user to redirect e-mails intended for a certain user to the account of another user or multiple users. It can also be used to implement the domain alias mechanism.

Examples of the entry in the file:

Redirecting email for user1, to root users and user3:

user1 root, user3

Redirecting email for user 1 in the example.com domain to the root user:

user1@example.com root

After you make change to this file, you must also run the postmap command:

# postmap /etc/postfix/virtual
# ll /etc/postfix/virtual

-rw-r\--r\--. 1 root root 12494 2014-06-10 /etc/postfix/virtual
-rw-r\--r\--. 1 root root 12288 07-31 21:58 /etc/postfix/virtual.db

Basic postfix configuration

Base configuration of postfix application you can make in /etc/postfix/main.cfg configuration file, which must complete with the following entry:

• section # RECEIVING MAIL

    inet_interfaces = all
    inet_protocols = ipv4
  

• section # INTERNET OR INTRANET

    relayhost = [IP mail server]:25 (port number)
  

I the netxt step you must complete the canonical file of postfix

At the end you should restart the postfix:

systemctl restart postfix

Example of postfix configuration with SSL encryption enabled

To configure email delivery with SSL encryption you need to make the following changes in the postfix configuration files:

• /etc/postfix/main.cf - file should contain the following entries in addition to standard (unchecked entries):

    mydestination = $myhostname, localhost.$mydomain, localhost
    myhostname = example.com
    relayhost = [smtp.example.com]:587
    smtp_sasl_auth_enable = yes
    smtp_sasl_password_maps = hash:/etc/postfix/sasl_passwd
    smtp_sasl_security_options = noanonymous
    smtp_tls_CAfile = /root/certs/cacert.cer
    smtp_use_tls = yes
    smtp_sasl_mechanism_filter = plain, login
    smtp_sasl_tls_security_options = noanonymous
    canonical_maps = hash:/etc/postfix/canonical
    smtp_generic_maps = hash:/etc/postfix/generic
    smtpd_recipient_restrictions = permit_sasl_authenticated
  

• /etc/postfix/sasl/passwd - file should define the data for authorized

   	[smtp.example.com\]:587 [[USER@example.com:PASS]](mailto:USER@example.com:PASS)
  

You need to give appropriate permissions:

	chmod 400 /etc/postfix/sasl_passwd

and map configuration to database:

	postmap /etc/postfix/sasl_passwd

next you need to generate a ca cert file:

	cat /etc/ssl/certs/Example\_Server\_CA.pem | tee -a etc/postfix/cacert.pem

And finally, you need to restart postfix

	/etc/init.d/postfix restart

API

Elasticsearch API

The Elasticsearch has a typical REST API and data is received in JSON format after the HTTP protocol. By default the tcp/9200 port is used to communicate with the Elasticsearch API. For purposes of examples, communication with the Elasticsearch API will be carried out using the curl application.

Program syntax:

# curl -XGET -u login:password '127.0.0.1:9200'

Available methods:

• PUT - sends data to the server;
• POST - sends a request to the server for a change;
• DELETE - deletes the index / document;
• GET - gets information about the index /document;
• HEAD - is used to check if the index / document exists.

Avilable APIs by roles:

• Index API - manages indexes;
• Document API - manges documnets;
• Cluster API - manage the cluster;
• Search API - is userd to search for data.

Elasticsearch Index API

The indices APIs are used to manage individual indices, index settings, aliases, mappings, and index templates.

Adding Index

Adding Index - autormatic method:

# curl -XPUT -u login:password '127.0.0.1:9200/twitter/tweet/1?pretty=true' -d'{
    "user" : "elk01",
    "post_date" : "2017-09-05T10:00:00",
    "message" : "tests auto index generation"
   }'

You should see the output:

{
"_index" : "twitter",
  "_type" : "tweet",
  "_id" : "1",
  "_version" : 1,
  "_shards" : {
    "total" : 2,
    "successful" : 1,
    "failed" : 0
  },
  "created" : true
}

The parameter action.auto_create_index must be set on true.

Adding Index – manual method:

• settings the number of shards and replicas:

    # curl -XPUT -u login:password '127.0.0.1:9200/twitter2?pretty=true' -d'{
    	"settings" : {
    	"number_of_shards" : 1,
    	"number_of_replicas" : 1
    	}
           }'
  

You should see the output:

{
  "acknowledged" : true
}

• command for manual index generation:

    # curl -XPUT -u login:password '127.0.0.1:9200/twitter2/tweet/1?pretty=true' -d'{
    			"user" : "elk01",
    			"post_date" : "2017-09-05T10:00:00",
    			"message" : "tests manual index generation"
    		}'
  

You should see the output:

{
  "_index" : "twitter2",
  "_type" : "tweet",
  "_id" : "1",
  "_version" : 1,
  "_shards" : {
    "total" : 2,
     "successful" : 1,
     "failed" : 0
  },
  "created" : true
}

Delete Index

Delete Index - to delete twitter index you need use the following command:

# curl -XDELETE -u login:password '127.0.0.1:9200/twitter?pretty=true'

The delete index API can also be applied to more than one index, by either using a comma separated list, or on all indice by using _all or * as index:

# curl -XDELETE -u login:password '127.0.0.1:9200/twitter*?pretty=true'

To allowing to delete indices via wildcards set action.destructive_requires_name setting in the config to false.

API useful commands

• get information about Replicas and Shards:

    # curl -XGET -u login:password '127.0.0.1:9200/twitter/_settings?pretty=true'
    # curl -XGET -u login:password '127.0.0.1:9200/twitter2/_settings?pretty=true'
  

• get information about mapping and alias in the index:

    # curl -XGET -u login:password '127.0.0.1:9200/twitter/_mappings?pretty=true'
    # curl -XGET -u login:password '127.0.0.1:9200/twitter/_aliases?pretty=true'
  

• get all information about the index:

    # curl -XGET -u login:password '127.0.0.1:9200/twitter?pretty=true'
  

• checking does the index exist:

    # curl -XGET -u login:password '127.0.0.1:9200/twitter?pretty=true'
  

• close the index:

    # curl -XGET -u login:password '127.0.0.1:9200/twitter/_close?pretty=true'
  

• open the index:

    # curl -XGET -u login:password '127.0.0.1:9200/twitter/_open?pretty=true'
  

• get the status of all indexes:

    # curl -XGET -u login:password '127.0.0.1:9200/_cat/indices?v'
  

• get the status of one specific index:

    # curl -XGET -u login:password '127.0.0.1:9200/_cat/indices/twitter?v'
  

• display how much memory is used by the indexes:

    # curl -XGET -u login:password '127.0.0.1:9200/_cat/indices?v&h=i,tm&s=tm:desc'
  

• display details of the shards:

    # curl -XGET -u login:password '127.0.0.1:9200/_cat/shards?v'# Elasticsearch Document API #
  

Elasticsearch Document API

Create Document

• create a document with a specify ID:

    # curl -XPUT -u login:password '127.0.0.1:9200/twitter/tweet/1?pretty=true' -d'{
    		"user" : "lab1",
    		"post_date" : "2017-08-25T10:00:00",
    		"message" : "testuje Elasticsearch"
    	}'
  

You should see the output:

{
  "_index" : "twitter",
  "_type" : "tweet",
  "_id" : "1",
  "_version" : 1,
  "_shards" : {
    "total" : 2,
    "successful" : 1,
    "failed" : 0
  },
  "created" : true
}

• creating a document with an automatically generated ID: (note: PUT-> POST):

    # curl -XPOST -u login:password '127.0.0.1:9200/twitter/tweet?pretty=true' -d'{
    		"user" : "lab1",
    		"post_date" : "2017-08-25T10:10:00",
    		"message" : "testuje automatyczne generowanie ID"
    		}'
  

You should see the output:

{
  "_index" : "twitter",
  "_type" : "tweet",
  "_id" : "AV49sTlM8NzerkV9qJfh",
  "_version" : 1,
  "_shards" : {
    "total" : 2,
    "successful" : 1,
    "failed" : 0
  },
  "created" : true
}

Delete Document

• delte a document by ID:

    # curl -XDELETE -u login:password '127.0.0.1:9200/twitter/tweet/1?pretty=true'
    # curl -XDELETE -u login:password '127.0.0.1:9200/twitter/tweet/AV49sTlM8NzerkV9qJfh?pretty=true'
  

• delete a document using a wildcard:

    # curl -XDELETE -u login:password '127.0.0.1:9200/twitter/tweet/1*?pretty=true'
  

(parametr: action.destructive_requires_name must be set to false)

Useful commands

• get information about the document:

    # curl -XGET -u login:password '127.0.0.1:9200/twitter/tweet/1?pretty=true'
  

You should see the output:

{
  "_index" : "twitter",
  "_type" : "tweet",
  "_id" : "1",
  "_version" : 1,
  "found" : true,
  "_source" : {
    "user" : "lab1",
    "post_date" : "2017-08-25T10:00:00",
    "message" : "testuje Elasticsearch"
  }
}

• get the source of the document:

    # curl -XGET -u login:password '127.0.0.1:9200/twitter/tweet/1/_source?pretty=true'
  

You should see the output:

{
  "user" : "lab1",
  "post_date" : "2017-08-25T10:00:00",
  "message" : "test of Elasticsearch"
}

• get information about all documents in the index:

    # curl -XGET -u login:password '127.0.0.1:9200/twitter*/_search?q=*&pretty=true'
  

You should see the output:

{
  "took" : 7,
  "timed_out" : false,
  "_shards" : {
    "total" : 10,
    "successful" : 10,
    "failed" : 0
  },
  "hits" : {
    "total" : 3,
    "max_score" : 1.0,
    "hits" : [ {
      "_index" : "twitter",
      "_type" : "tweet",
      "_id" : "AV49sTlM8NzerkV9qJfh",
      "_score" : 1.0,
      "_source" : {
        "user" : "lab1",
        "post_date" : "2017-08-25T10:10:00",
        "message" : "auto generated ID"
      }
    }, {
      "_index" : "twitter",
      "_type" : "tweet",
      "_id" : "1",
      "_score" : 1.0,
      "_source" : {
        "user" : "lab1",
        "post_date" : "2017-08-25T10:00:00",
        "message" : "Elasticsearch test"
      }
    }, {
      "_index" : "twitter2",
      "_type" : "tweet",
      "_id" : "1",
      "_score" : 1.0,
      "_source" : {
        "user" : "elk01",
        "post_date" : "2017-09-05T10:00:00",
        "message" : "manual index created test"
      }
    } ]
  }
}

• the sum of all documents in a specified index:

    # curl -XGET -u login:password '127.0.0.1:9200/_cat/count/twitter?v'
  

You should see the output:

	epoch      		timestamp count
	1504281400 	17:56:40     2

• the sum of all document in Elasticsearch database:

    # curl -XGET -u login:password '127.0.0.1:9200/_cat/count?v'
  

You should see the output:

	epoch      		timestamp count
	1504281518 	17:58:38    493658

Elasticsearch Cluster API

Useful commands

• information about the cluster state:

    # curl -XGET -u login:password '127.0.0.1:9200/_cluster/health?pretty=true'
  

• information about the role and load of nodes in the cluster:

    # curl -XGET -u login:password '127.0.0.1:9200/_cat/nodes?v'
  

• information about the available and used place on the cluster nodes:

    # curl -XGET -u login:password '127.0.0.1:9200/_cat/allocation?v'
  

• information which node is currently in the master role:

    # curl -XGET -u login:password '127.0.0.1:9200/_cat/master?v'
  

• information abut currently performed operations by the cluster:

    # curl -XGET -u login:password '127.0.0.1:9200/_cat/pending_tasks?v'
  

• information on revoceries / transferred indices:

    # curl -XGET -u login:password '127.0.0.1:9200/_cat/recovery?v'
  

• information about shards in a cluster:

    # curl -XGET -u login:password '127.0.0.1:9200/_cat/shards?v'
  

• detailed inforamtion about the cluster:

    # curl -XGET -u login:password '127.0.0.1:9200/_cluster/stats?human&pretty'
  

• detailed information about the nodes:

    # curl -XGET -u login:password '127.0.0.1:9200/_nodes/stats?human&pretty'
  

Elasticsearch Search API

Useful commands

• searching for documents by the string:

    # curl -XPOST -u login:password '127.0.0.1:9200/twitter*/tweet/_search?pretty=true' -d '{
    			"query": {
        				"bool" : {
            					"must" : {
                					"query_string" : {
                    						"query" : "test"
                					}
            					}
    			}		
   			 }
    	}'
  

• searching for document by the string and filtering:

    # curl -XPOST -u login:password '127.0.0.1:9200/twitter*/tweet/_search?pretty=true' -d'{
    		"query": {
    				"bool" : {
        					"must" : {
            					"query_string" : {
                						"query" : "testuje"
            						}
        						},
        					"filter" : {
            					"term" : { "user" : "lab1" }
        					}
    				}
    		}
    }'
  

• simple search in a specific field (in this case user) uri query:

    	# curl -XGET -u login:password '127.0.0.1:9200/twitter*/_search?q=user:lab1&pretty=true'
  

• simple search in a specific field:

    # curl -XPOST -u login:password '127.0.0.1:9200/twitter*/_search?pretty=true' -d '{
    			 "query" : {
           				 "term" : { "user" : "lab1" }
      	  		}
    		}'
  

Elasticsearch - Mapping, Fielddata and Templates

Mapping is a collection of fields along with a specific data type Fielddata is the field in which the data is stored (requires a specific type - string, float) Template is a template based on which fielddata will be created in a given index.

Useful commands

• Information on all set mappings:

    # curl -XGET -u login:password '127.0.0.1:9200/_mapping?pretty=true'
  

• Information about all mappings set in the index:

    # curl -XGET -u login:password '127.0.0.1:9200/twitter/_mapping/*?pretty=true'
  

• Information about the type of a specific field:

    # curl -XGET -u login:password '127.0.0.1:9200/twitter/_mapping/field/message*?pretty=true'
  

• Information on all set templates:

    # curl  -XGET -u login:password '127.0.0.1:9200/_template/*?pretty=true'
  

Create - Mapping / Fielddata

• Create - Mapping / Fielddata - It creates index twitter-float and the tweet message field sets to float:

     # curl -XPUT -u login:password '127.0.0.1:9200/twitter-float?pretty=true' -d '{
       "mappings": {
         "tweet": {
           "properties": {
             "message": {
               "type": "float"
             }
           }
         }
       }
     }'
  
    # curl -XGET -u login:password '127.0.0.1:9200/twitter-float/_mapping/field/message?pretty=true'
  

Create Template

• Create Template:

    # curl -XPUT -u login:password '127.0.0.1:9200/_template/template_1' -d'{
        "template" : "twitter4",
        "order" : 0,
        "settings" : {
            "number_of_shards" : 2
        }
    }'
  
    # curl -XPOST -u login:password '127.0.0.1:9200/twitter4/tweet?pretty=true' -d'{
    "user" : "lab1",
    "post_date" : "2017-08-25T10:10:00",
    "message" : "test of ID generation"
    }'
  
    # curl -XGET -u login:password '127.0.0.1:9200/twitter4/_settings?pretty=true'
  

• Create Template2 - Sets the mapping template for all new indexes specifying that the tweet data, in the field called message, should be of the “string” type:

    # curl -XPUT -u login:password '127.0.0.1:9200/_template/template_2' -d'{
      "template" : "*",
      "mappings": {
        "tweet": {
          "properties": {
            "message": {
              "type": "string"
            }
          }
        }
      }
    }'
  

Delete Mapping

• Delete Mapping - Deleting a specific index mapping (no possibility to delete - you need to index):

    # curl -XDELETE -u login:password '127.0.0.1:9200/twitter2'
  

Delete Template

• Delete Template:

    # curl  -XDELETE -u login:password '127.0.0.1:9200/_template/template_1?pretty=true'
  

AI Module API

Services

The intelligence module has implemented services that allow you to create, modify, delete, execute and read definitions of AI rules.

List rules

The list service returns a list of AI rules definitions stored in the system.

Method: GET URL:

https://<host>:<port>/api/ai/list?pretty

where:

host	-	kibana host address
port	-	kibana port
?pretty	-	optional json format parameter

Curl:

curl -XGET 'https://logserver.test.pl:5601/api/ai/list?pretty' -u <user>:<password> -k 

Result: Array of JSON documents:

| Field                          	| Value                                                                               	| Screen field (description) 	|
|--------------------------------	|-------------------------------------------------------------------------------------	|----------------------------	|
| _source.algorithm_type         	| GMA, GMAL, LRS, LRST, RFRS, SMAL, SMA, TL                                           	| Algorithm.                 	|
| _source.model_name             	| Not empty string.                                                                   	| AI Rule Name.              	|
| _source.search                 	| Search id.                                                                          	| Choose search.             	|
| _source.label_field.field      	|                                                                                     	| Feature to analyse.        	|
| _source.max_probes             	| Integer value                                                                       	| Max probes                 	|
| _source.time_frame             	| 1 minute, 5 minutes, 15 minutes, 30 minutes, 1 hour, 1 day, 1 week, 30 day, 365 day 	| Time frame                 	|
| _source.value_type             	| min, max, avg, count                                                                	| Value type                 	|
| _source.max_predictions        	| Integer value                                                                       	| Max predictions            	|
| _source.threshold              	| Integer value                                                                       	| Threshold                  	|
| _source.automatic_cron         	| Cron format string                                                                  	| Automatic cycle            	|
| _source.automatic_enable       	| true/false                                                                          	| Enable                     	|
| _source.automatic              	| true/false                                                                          	| Automatic                  	|
| _source.start_date             	| YYYY-MM-DD HH:mm or now                                                             	| Start date                 	|
| _source.multiply_by_values     	| Array of string values                                                              	| Multiply by values         	|
| _source.multiply_by_field      	| None or full field name eg.: system.cpu                                             	| Multiply by field          	|
| _source.selectedroles          	| Array of roles name                                                                 	| Role                       	|
| _source.last_execute_timestamp 	|                                                                                     	| Last execute               	|

Not screen fields:

| _index                        	|   	| Elasticsearch index name.           	|
|-------------------------------	|---	|-------------------------------------	|
| _type                         	|   	| Elasticsearch document type.        	|
| _id                           	|   	| Elasticsearch document id.          	|
| _source.preparation_date      	|   	| Document preparation date.          	|
| _source.machine_state_uid     	|   	| AI rule machine state uid.          	|
| _source.path_to_logs          	|   	| Path to ai machine logs.            	|
| _source.path_to_machine_state 	|   	| Path to ai machine state files.     	|
| _source.searchSourceJSON      	|   	| Query string.                       	|
| _source.processing_time       	|   	| Process operation time.             	|
| _source.last_execute_mili     	|   	| Last executed time in milliseconds. 	|
| _source.pid                   	|   	| Process pid if ai rule is running.  	|
| _source.exit_code             	|   	| Last executed process exit code.    	|

Show rules

The show service returns a document of AI rule definition by id.

Method: GET URL: https://:/api/ai/show/?pretty

where:

host	-	kibana host address
port	-	kibana port
id	-	ai rule document id
?pretty	-	optional json format parameter

Curl:

curl -XGET 'https://logserver.test.pl:5601/api/ai/show/ea9384857de1f493fd84dabb6dfb99ce?pretty' -u <user>:<password> -k

Result JSON document:

| Field                          	| Value                                                                               	| Screen field (description) 	|
|--------------------------------	|-------------------------------------------------------------------------------------	|----------------------------	|
| _source.algorithm_type         	| GMA, GMAL, LRS, LRST, RFRS, SMAL, SMA, TL                                           	| Algorithm.                 	|
| _source.model_name             	| Not empty string.                                                                   	| AI Rule Name.              	|
| _source.search                 	| Search id.                                                                          	| Choose search.             	|
| _source.label_field.field      	|                                                                                     	| Feature to analyse.        	|
| _source.max_probes             	| Integer value                                                                       	| Max probes                 	|
| _source.time_frame             	| 1 minute, 5 minutes, 15 minutes, 30 minutes, 1 hour, 1 day, 1 week, 30 day, 365 day 	| Time frame                 	|
| _source.value_type             	| min, max, avg, count                                                                	| Value type                 	|
| _source.max_predictions        	| Integer value                                                                       	| Max predictions            	|
| _source.threshold              	| Integer value                                                                       	| Threshold                  	|
| _source.automatic_cron         	| Cron format string                                                                  	| Automatic cycle            	|
| _source.automatic_enable       	| true/false                                                                          	| Enable                     	|
| _source.automatic              	| true/false                                                                          	| Automatic                  	|
| _source.start_date             	| YYYY-MM-DD HH:mm or now                                                             	| Start date                 	|
| _source.multiply_by_values     	| Array of string values                                                              	| Multiply by values         	|
| _source.multiply_by_field      	| None or full field name eg.: system.cpu                                             	| Multiply by field          	|
| _source.selectedroles          	| Array of roles name                                                                 	| Role                       	|
| _source.last_execute_timestamp 	|                                                                                     	| Last execute               	|

Not screen fields

| _index                        	|   	| Elasticsearch index name.           	|
|-------------------------------	|---	|-------------------------------------	|
| _type                         	|   	| Elasticsearch document type.        	|
| _id                           	|   	| Elasticsearch document id.          	|
| _source.preparation_date      	|   	| Document preparation date.          	|
| _source.machine_state_uid     	|   	| AI rule machine state uid.          	|
| _source.path_to_logs          	|   	| Path to ai machine logs.            	|
| _source.path_to_machine_state 	|   	| Path to ai machine state files.     	|
| _source.searchSourceJSON      	|   	| Query string.                       	|
| _source.processing_time       	|   	| Process operation time.             	|
| _source.last_execute_mili     	|   	| Last executed time in milliseconds. 	|
| _source.pid                   	|   	| Process pid if ai rule is running.  	|
| _source.exit_code             	|   	| Last executed process exit code.    	|

Create rules

The create service adds a new document with the AI rule definition.

Method: PUT

URL:

https://<host>:<port>/api/ai/create

where:

host	-	kibana host address
port	-	kibana port
body	-	JSON with definition of ai rule

Curl:

curl -XPUT 'https://logserver.test.pl:5601/api/ai/create' -u <user>:<password> -k -H "kbn-version: 6.2.4" -H 'Content-type: application/json' -d' {"algorithm_type":"TL","model_name":"test","search":"search:6c226420-3b26-11e9-a1c0-4175602ff5d0","label_field":{"field":"system.cpu.idle.pct"},"max_probes":100,"time_frame":"1 day","value_type":"avg","max_predictions":10,"threshold":-1,"automatic_cron":"*/5 * * * *","automatic_enable":true,"automatic_flag":true,"start_date":"now","multiply_by_values":[],"multiply_by_field":"none","selectedroles":["test"]}'

Validation:

| Field          	| Values                                                                              	|
|----------------	|-------------------------------------------------------------------------------------	|
| algorithm_type 	| GMA, GMAL, LRS, LRST, RFRS, SMAL, SMA, TL                                           	|
| value_type     	| min, max, avg, count                                                                	|
| time_frame     	| 1 minute, 5 minutes, 15 minutes, 30 minutes, 1 hour, 1 day, 1 week, 30 day, 365 day 	|

Body JSON description:

| Field              	| Mandatory        	| Value                                                                               	| Screen field        	|
|--------------------	|------------------	|-------------------------------------------------------------------------------------	|---------------------	|
| algorithm_type     	| Yes              	| GMA, GMAL, LRS, LRST, RFRS, SMAL, SMA, TL                                           	| Algorithm.          	|
| model_name         	| Yes              	| Not empty string.                                                                   	| AI Rule Name.       	|
| search             	| Yes              	| Search id.                                                                          	| Choose search.      	|
| label_field.field  	| Yes              	|                                                                                     	| Feature to analyse. 	|
| max_probes         	| Yes              	| Integer value                                                                       	| Max probes          	|
| time_frame         	| Yes              	| 1 minute, 5 minutes, 15 minutes, 30 minutes, 1 hour, 1 day, 1 week, 30 day, 365 day 	| Time frame          	|
| value_type         	| Yes              	| min, max, avg, count                                                                	| Value type          	|
| max_predictions    	| Yes              	| Integer value                                                                       	| Max predictions     	|
| threshold          	| No (default -1)  	| Integer value                                                                       	| Threshold           	|
| automatic_cron     	| Yes              	| Cron format string                                                                  	| Automatic cycle     	|
| Automatic_enable   	| Yes              	| true/false                                                                          	| Enable              	|
| automatic          	| Yes              	| true/false                                                                          	| Automatic           	|
| start_date         	| No (default now) 	| YYYY-MM-DD HH:mm or now                                                             	| Start date          	|
| multiply_by_values 	| Yes              	| Array of string values                                                              	| Multiply by values  	|
| multiply_by_field  	| Yes              	| None or full field name eg.: system.cpu                                             	| Multiply by field   	|
| selectedroles      	| No               	| Array of roles name                                                                 	| Role                	|

Result:

JSON document with fields:

status	-	true if ok
id	-	id of changed document
message	-	error message

Update rules

The update service changes the document with the AI rule definition.

Method:POST

URL:

https://<host>:<port>/api/ai/update/<id>

where:

host	-	kibana host address
port	-	kibana port
id	-	ai rule document id
body	-	JSON with definition of ai rule

Curl:

curl -XPOST 'https://logserver.test.pl:5601/api/ai/update/ea9384857de1f493fd84dabb6dfb99ce' -u <user>:<password> -k -H "kbn-version: 6.2.4" -H 'Content-type: application/json' -d'
{"algorithm_type":"TL","search":"search:6c226420-3b26-11e9-a1c0-4175602ff5d0","label_field":{"field":"system.cpu.idle.pct"},"max_probes":100,"time_frame":"1 day","value_type":"avg","max_predictions":100,"threshold":-1,"automatic_cron":"*/5 * * * *","automatic_enable":true,"automatic_flag":true,"start_date":"now","multiply_by_values":[],"multiply_by_field":"none","selectedroles":["test"]}

Validation:

| Field          	| Values                                                                              	|
|----------------	|-------------------------------------------------------------------------------------	|
| algorithm_type 	| GMA, GMAL, LRS, LRST, RFRS, SMAL, SMA, TL                                           	|
| value_type     	| min, max, avg, count                                                                	|
| time_frame     	| 1 minute, 5 minutes, 15 minutes, 30 minutes, 1 hour, 1 day, 1 week, 30 day, 365 day 	|

Body JSON description:

| Field              	| Mandatory        	| Value                                                                               	| Screen field        	|
|--------------------	|------------------	|-------------------------------------------------------------------------------------	|---------------------	|
| algorithm_type     	| Yes              	| GMA, GMAL, LRS, LRST, RFRS, SMAL, SMA, TL                                           	| Algorithm.          	|
| model_name         	| Yes              	| Not empty string.                                                                   	| AI Rule Name.       	|
| search             	| Yes              	| Search id.                                                                          	| Choose search.      	|
| label_field.field  	| Yes              	|                                                                                     	| Feature to analyse. 	|
| max_probes         	| Yes              	| Integer value                                                                       	| Max probes          	|
| time_frame         	| Yes              	| 1 minute, 5 minutes, 15 minutes, 30 minutes, 1 hour, 1 day, 1 week, 30 day, 365 day 	| Time frame          	|
| value_type         	| Yes              	| min, max, avg, count                                                                	| Value type          	|
| max_predictions    	| Yes              	| Integer value                                                                       	| Max predictions     	|
| threshold          	| No (default -1)  	| Integer value                                                                       	| Threshold           	|
| automatic_cron     	| Yes              	| Cron format string                                                                  	| Automatic cycle     	|
| Automatic_enable   	| Yes              	| true/false                                                                          	| Enable              	|
| automatic          	| Yes              	| true/false                                                                          	| Automatic           	|
| start_date         	| No (default now) 	| YYYY-MM-DD HH:mm or now                                                             	| Start date          	|
| multiply_by_values 	| Yes              	| Array of string values                                                              	| Multiply by values  	|
| multiply_by_field  	| Yes              	| None or full field name eg.: system.cpu                                             	| Multiply by field   	|
| selectedroles      	| No               	| Array of roles name                                                                 	| Role                	|

Result:

JSON document with fields:

	status	-	true if ok
	id	-	id of changed document
	message	-	error message

Run:

The run service executes a document of AI rule definition by id.

Method: GET

URL:

https://<host>:<port>/api/ai/run/<id>

where:

	host	-	kibana host address
	port	-	kibana port
	id	-	ai rule document id

Curl:

	curl -XGET 'https://logserver.test.pl:5601/api/ai/run/ea9384857de1f493fd84dabb6dfb99ce' -u <user>:<password> -k

Result:

JSON document with fields:

	status	-	true if ok
	id	-	id of executed document
	message	-	message

Delete rules

The delete service removes a document of AI rule definition by id.

Method: DELETE

URL:

https://<host>:<port>/api/ai/delete/<id>

where:

	host	-	kibana host address
	port	-	kibana port
	id	-	ai rule document id

Curl:

curl -XDELETE 'https://logserver.test.pl:5601/api/ai/delete/ea9384857de1f493fd84dabb6dfb99ce' -u <user>:<password> -k -H "kbn-version: 6.2.4"

Result:

JSON document with fields:

status	-	true if ok
id	-	id of executed document
message	-	message

Alert module API

Create Alert Rule

Method: POST

URL: /api/admin/alertrules

Body:

In the body of call, you must pass the JSON object with the full definition of the rule document:

| Name                  | Description                                                                                                                                                                      |
|-----------------------|----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
| id                    | Document ID in Elasticsearch                                                                                                                                                     |
| alertrulename         | Rule name (the Name field from the Create Alert tab  the name must be the same as the alert name)                                                                                |
| alertruleindexpattern | Index pattern (Index pattern field from the Create Alert tab)                                                                                                                    |
| selectedroles         | Array of roles that have rights to this rule (Roles field from the Create Alert tab)                                                                                             |
| alertruletype         | Alert rule type (Type field from the Create Alert tab)                                                                                                                           |
| alertrulemethod       | Type of alert method (Alert method field from the Create Alert tab)                                                                                                              |
| alertrulemethoddata   | Data for the type of alert (field Email address if alertrulemethod is email  Path to script / command if alertrulemethod is command  and empty value if alertrulemethod is none) |
| alertrule_any         | Alert script (the Any field from the Create Alert tab)                                                                                                                           |
| alertruleimportance   | Importance of the rule (Rule importance box from the Create Alert tab)                                                                                                           |
| alertruleriskkey      | Field for risk calculation (field from the index indicated by alertruleindexpattern according to which the risk will be counted  Risk key field from the Create Alert tab)       |
| alertruleplaybooks    | Playbook table (document IDs) attached to the alert (Playbooks field from the Create Alert tab)                                                                                  |
| enable                | Value Y or N depending on whether we enable or disable the rule                                                                                                                  |
| authenticator         | Constant value index                                                                                                                                                             |

Result OK:

 "Successfully created rule!!" 

or if fault, error message.

Example:

curl -XPOST 'https://localhost:5601/api/admin/alertrules' -u logserver:logserver -k -H "kbn-version: 6.2.4" -H 'Content-type: application/json' -d'
{
	"id":"test_enable_rest",
	"alertrulename":"test enable rest",
	"alertruleindexpattern":"m*",
	"selectedroles":"",
	"alertruletype":"frequency",
	"alertrulemethod":"email",
	"alertrulemethoddata":"ala@local",
	"alertrule_any":"# (Required, frequency specific)\n# Alert when this many documents matching the query occur within a timeframe\nnum_events: 5\n\n# (Required, frequency specific)\n# num_events must occur within this amount of time to trigger an alert\ntimeframe:\n  minutes: 2\n\n# (Required)\n# A list of Elasticsearch filters used for find events\n# These filters are joined with AND and nested in a filtered query\n# For more info: http://www.elasticsearch.org/guide/en/elasticsearch/reference/current/query-dsl.html\nfilter:\n- term:\n    some_field: \"some_value\"\n\n# (Optional, change specific)\n# If true, Alert will poll Elasticsearch using the count api, and not download all of the matching documents. This is useful is you care only about numbers and not the actual data. It should also be used if you expect a large number of query hits, in the order of tens of thousands or more. doc_type must be set to use this.\n#use_count_query:\n\n# (Optional, change specific)\n# Specify the _type of document to search for. This must be present if use_count_query or use_terms_query is set.\n#doc_type:\n\n# (Optional, change specific)\n# If true, Alert will make an aggregation query against Elasticsearch to get counts of documents matching each unique value of query_key. This must be used with query_key and doc_type. This will only return a maximum of terms_size, default 50, unique terms.\n#use_terms_query:\n\n# (Optional, change specific)\n# When used with use_terms_query, this is the maximum number of terms returned per query. Default is 50.\n#terms_size:\n\n# (Optional, change specific)\n# Counts of documents will be stored independently for each value of query_key. Only num_events documents, all with the same value of query_key, will trigger an alert.\n#query_key:\n\n# (Optional, change specific)\n# Will attach all the related events to the event that triggered the frequency alert. For example in an alert triggered with num_events: 3, the 3rd event will trigger the alert on itself and add the other 2 events in a key named related_events that can be accessed in the alerter.\n#attach_related:",
	"alertruleplaybooks":[],
	"alertruleimportance":50,
	"alertruleriskkey":"beat.hostname",
	"enable":"Y",
	"authenticator":"index"
}
'

Save Alert Rules

Method: POST

URL:

/api/admin/saverules

Body:

In the body of call, you must pass the JSON object:

'authenticator'

Constant value index

Result:

"Files created" 

or if fault, error message.

Example:

curl -XPOST 'https://localhost:5601/api/admin/saverules' -u logserver:logserver -k -H "kbn-version: 6.2.4" -H 'Content-type: application/json' -d'
	{
		"authenticator":"index"
	}
'

Logstash

The OP5 Log Analytics use Logstash service to dynamically unify data from disparate sources and normalize the data into destination of your choose. A Logstash pipeline has two required elements, input and output, and one optional element filter. The input plugins consume data from a source, the filter plugins modify the data as you specify, and the output plugins write the data to a destination. The default location of the Logstash plugin files is: /etc/logstash/conf.d/. This location contain following OP5

Log Analytics default plugins:

• 01-input-beats.conf
• 01-input-syslog.conf
• 020-filter-beats-syslog.conf
• 020-filter-network.conf
• 099-filter-geoip.conf
• 100-output-elasticsearch.conf
• naemon_beat.example
• perflogs.example

Logstash - Input “beats”

This plugin wait for receiving data from remote beats services. It use tcp /5044 port for communication:

            input {
                    beats {
                            port => 5044
                    }
            }

Logstash - Input “network”

This plugin read events over a TCP or UDP socket assigns the appropriate tags:

	input {
	        tcp {
	                port => 5514
	                type => "network"
	
	                tags => [ "LAN", "TCP" ]
	        }
	
	        udp {
	                port => 5514
	                type => "network"
	
	                tags => [ "LAN", "UDP" ]
	        }
	}

Logstash - Filter “beats syslog”

This filter processing an event data with syslog type:

	filter {
	
	 if [type] == "syslog" {
			grok {
					match => { 
					  "message" => [
					# auth: ssh|sudo|su
	
					  "%{SYSLOGTIMESTAMP:[system][auth][timestamp]} %{SYSLOGHOST:[system][auth][hostname]} sshd(?:\[%{POSINT:[system][auth][pid]}\])?: %{DATA:[system][auth][ssh][event]} %{DATA:[system][auth][ssh][method]} for (invalid user )?%{DATA:[system][auth][user]} from %{IPORHOST:[system][auth][ssh][ip]} port %{NUMBER:[system][auth][ssh][port]} ssh2(: %{GREEDYDATA:[system][auth][ssh][signature]})?",
	
							  "%{SYSLOGTIMESTAMP:[system][auth][timestamp]} %{SYSLOGHOST:[system][auth][hostname]} sshd(?:\[%{POSINT:[system][auth][pid]}\])?: %{DATA:[system][auth][ssh][event]} user %{DATA:[system][auth][user]} from %{IPORHOST:[system][auth][ssh][ip]}",
	
							  "%{SYSLOGTIMESTAMP:[system][auth][timestamp]} %{SYSLOGHOST:[system][auth][hostname]} sshd(?:\[%{POSINT:[system][auth][pid]}\])?: Did not receive identification string from %{IPORHOST:[system][auth][ssh][dropped_ip]}",
	
							  "%{SYSLOGTIMESTAMP:[system][auth][timestamp]} %{SYSLOGHOST:[system][auth][hostname]} sudo(?:\[%{POSINT:[system][auth][pid]}\])?: \s*%{DATA:[system][auth][user]} :( %{DATA:[system][auth][sudo][error]} ;)? TTY=%{DATA:[system][auth][sudo][tty]} ; PWD=%{DATA:[system][auth][sudo][pwd]} ; USER=%{DATA:[system][auth][sudo][user]} ; COMMAND=%{GREEDYDATA:[system][auth][sudo][command]}",
	
							  "%{SYSLOGTIMESTAMP:[system][auth][timestamp]} %{SYSLOGHOST:[system][auth][hostname]} %{DATA:[system][auth][program]}(?:\[%{POSINT:[system][auth][pid]}\])?: %{GREEDYMULTILINE:[system][auth][message]}",
	
					# add/remove user or group
							  "%{SYSLOGTIMESTAMP:[system][auth][timestamp]} %{SYSLOGHOST:[system][auth][hostname]} groupadd(?:\[%{POSINT:[system][auth][pid]}\])?: new group: name=%{DATA:system.auth.groupadd.name}, GID=%{NUMBER:system.auth.groupadd.gid}",
							  
					  "%{SYSLOGTIMESTAMP:[system][auth][timestamp]} %{SYSLOGHOST:[system][auth][hostname]} userdel(?:\[%{POSINT:[system][auth][pid]}\])?: removed group '%{DATA:[system][auth][groupdel][name]}' owned by '%{DATA:[system][auth][group][owner]}'",
	
							  "%{SYSLOGTIMESTAMP:[system][auth][timestamp]} %{SYSLOGHOST:[system][auth][hostname]} useradd(?:\[%{POSINT:[system][auth][pid]}\])?: new user: name=%{DATA:[system][auth][user][add][name]}, UID=%{NUMBER:[system][auth][user][add][uid]}, GID=%{NUMBER:[system][auth][user][add][gid]}, home=%{DATA:[system][auth][user][add][home]}, shell=%{DATA:[system][auth][user][add][shell]}$",
							  
					   "%{SYSLOGTIMESTAMP:[system][auth][timestamp]} %{SYSLOGHOST:[system][auth][hostname]} userdel(?:\[%{POSINT:[system][auth][pid]}\])?: delete user '%{WORD:[system][auth][user][del][name]}'$",
	
					   "%{SYSLOGTIMESTAMP:[system][auth][timestamp]} %{SYSLOGHOST:[system][auth][hostname]} usermod(?:\[%{POSINT:[system][auth][pid]}\])?: add '%{WORD:[system][auth][user][name]}' to group '%{WORD:[system][auth][user][memberof]}'",
	
					   # yum install/erase/update package
					   "%{SYSLOGTIMESTAMP:[system][auth][timestamp]} %{DATA:[system][package][action]}: %{NOTSPACE:[system][package][name]}"
					]
	
				}
				  
					pattern_definitions => {
					  "GREEDYMULTILINE"=> "(.|\n)*"
					}
				  }
	
				date {
						match => [ "[system][auth][timestamp]", 
						"MMM  d HH:mm:ss", 
					"MMM dd HH:mm:ss" 
					]
					target => "[system][auth][timestamp]"
				}
	
				mutate {
				  convert => { "[system][auth][pid]" => "integer" }
				  convert => { "[system][auth][groupadd][gid]" => "integer" }
				  convert => { "[system][auth][user][add][uid]" => "integer" }
				  convert => { "[system][auth][user][add][gid]" => "integer" }
				}
	  }
	}

Logstash Filter “network”

This filter processing an event data with network type:

filter {
 if [type] == "network" {
     grok {
		named_captures_only => true
		match => {
			"message" => [

			# Cisco Firewall
			"%{SYSLOG5424PRI}%{NUMBER:log_sequence#}:%{SPACE}%{IPORHOST:device_ip}: (?:.)?%{CISCOTIMESTAMP:log_data} CET: %%{CISCO_REASON:facility}-%{INT:severity_level}-%{CISCO_REASON:facility_mnemonic}:%{SPACE}%{GREEDYDATA:event_message}",

			# Cisco Routers
			"%{SYSLOG5424PRI}%{NUMBER:log_sequence#}:%{SPACE}%{IPORHOST:device_ip}: (?:.)?%{CISCOTIMESTAMP:log_data} CET: %%{CISCO_REASON:facility}-%{INT:severity_level}-%{CISCO_REASON:facility_mnemonic}:%{SPACE}%{GREEDYDATA:event_message}",

			# Cisco Switches
			"%{SYSLOG5424PRI}%{NUMBER:log_sequence#}:%{SPACE}%{IPORHOST:device_ip}: (?:.)?%{CISCOTIMESTAMP:log_data} CET: %%{CISCO_REASON:facility}-%{INT:severity_level}-%{CISCO_REASON:facility_mnemonic}:%{SPACE}%{GREEDYDATA:event_message}",
			"%{SYSLOG5424PRI}%{NUMBER:log_sequence#}:%{SPACE}(?:.)?%{CISCOTIMESTAMP:log_data} CET: %%{CISCO_REASON:facility}-%{INT:severity_level}-%{CISCO_REASON:facility_mnemonic}:%{SPACE}%{GREEDYDATA:event_message}",

			# HP switches
			"%{SYSLOG5424PRI}%{SPACE}%{CISCOTIMESTAMP:log_data} %{IPORHOST:device_ip} %{CISCO_REASON:facility}:%{SPACE}%{GREEDYDATA:event_message}"
			]

		}
	}

	syslog_pri { }

	if [severity_level] {

	  translate {
	    dictionary_path => "/etc/logstash/dictionaries/cisco_syslog_severity.yml"
	    field => "severity_level"
	    destination => "severity_level_descr"
	  }

	}

	if [facility] {

	  translate {
	    dictionary_path => "/etc/logstash/dictionaries/cisco_syslog_facility.yml"
	    field => "facility"
	    destination => "facility_full_descr"
	  }

	}

	 #ACL 
	 if [event_message] =~ /(\d+.\d+.\d+.\d+)/ {
	  grok {
	    match => {
		"event_message" => [
			"list %{NOTSPACE:[acl][name]} %{WORD:[acl][action]} %{WORD:[acl][proto]} %{IP:[src][ip]}.*%{IP:[dst][ip]}",
			"list %{NOTSPACE:[acl][name]} %{WORD:[acl][action]} %{IP:[src][ip]}",
			"^list %{NOTSPACE:[acl][name]} %{WORD:[acl][action]} %{WORD:[acl][proto]} %{IP:[src][ip]}.*%{IP:[dst][ip]}"
			]
	    }
	  }
	}

	if [src][ip] {

		cidr {
		   address => [ "%{[src][ip]}" ]
		   network => [ "0.0.0.0/32", "10.0.0.0/8", "172.16.0.0/12", "192.168.0.0/16", "fc00::/7", "127.0.0.0/8", "::1/128", "169.254.0.0/16", "fe80::/10","224.0.0.0/4", "ff00::/8","255.255.255.255/32"  ]
		   add_field => { "[src][locality]" => "private" }
		}

		if ![src][locality] {
		   mutate {
		      add_field => { "[src][locality]" => "public" }
		   }
		}
	}


	if [dst][ip] {
		cidr {
		   address => [ "%{[dst][ip]}" ]
		   network => [ "0.0.0.0/32", "10.0.0.0/8", "172.16.0.0/12", "192.168.0.0/16", "fc00::/7", "127.0.0.0/8", "::1/128",
				"169.254.0.0/16", "fe80::/10","224.0.0.0/4", "ff00::/8","255.255.255.255/32" ]
		   add_field => { "[dst][locality]" => "private" }
		}

		if ![dst][locality] {
		   mutate {
		      add_field => { "[dst][locality]" => "public" }
		   }
		}
	}

	# date format
	date {
	  match => [ "log_data",
		"MMM dd HH:mm:ss",
		"MMM  dd HH:mm:ss",
		"MMM dd HH:mm:ss.SSS",
		"MMM  dd HH:mm:ss.SSS",
		"ISO8601"
	  ]
	  target => "log_data"
	}

 }
}

Logstash - Filter “geoip”

This filter processing an events data with IP address and check localization:

filter {
     if [src][locality] == "public" {

		geoip {
			source => "[src][ip]"
			target => "[src][geoip]"
			database => "/etc/logstash/geoipdb/GeoLite2-City.mmdb"
			fields => [ "city_name", "country_name", "continent_code", "country_code2", "location" ]
			remove_field => [ "[src][geoip][ip]" ]
		}

		geoip {
			source => "[src][ip]"
			target => "[src][geoip]"
			database => "/etc/logstash/geoipdb/GeoLite2-ASN.mmdb"
			remove_field => [ "[src][geoip][ip]" ]
		}

     }

     if [dst][locality] == "public" {

		geoip {
			source => "[dst][ip]"
			target => "[dst][geoip]"
			database => "/etc/logstash/geoipdb/GeoLite2-City.mmdb"
			fields => [ "city_name", "country_name", "continent_code", "country_code2", "location" ]
			remove_field =>  [ "[dst][geoip][ip]" ]
		}

		geoip {
			source => "[dst][ip]"
			target => "[dst][geoip]"
			database => "/etc/logstash/geoipdb/GeoLite2-ASN.mmdb"
			remove_field => [ "[dst][geoip][ip]" ]
		}
     }

}

Logstash - Output to Elasticsearch

This output plugin sends all data to the local Elasticsearch instance and create indexes:

output {
	elasticsearch {
	   hosts => [ "127.0.0.1:9200" ]

	   index => "%{type}-%{+YYYY.MM.dd}"

	   user => "logstash"
	   password => "logstash"
	}
}

Logstash pluging for “naemon beat”

This Logstash plugin has example of complete configuration for integration with naemon application:

input {
    beats {
        port => FILEBEAT_PORT
        type => "naemon"
    }
}

filter {
    if [type] == "naemon" {
        grok {
            patterns_dir => [ "/etc/logstash/patterns" ]
            match => { "message" => "%{NAEMONLOGLINE}" }
            remove_field => [ "message" ]
        }
        date {
            match => [ "naemon_epoch", "UNIX" ]
            target => "@timestamp"
            remove_field => [ "naemon_epoch" ]
        }
    }
}

output {
    # Single index
#    if [type] == "naemon" {
#        elasticsearch {
#            hosts => ["ELASTICSEARCH_HOST:ES_PORT"]
#            index => "naemon-%{+YYYY.MM.dd}"
#        }
#    }

    # Separate indexes
    if [type] == "naemon" {
        if "_grokparsefailure" in [tags] {
            elasticsearch {
                hosts => ["ELASTICSEARCH_HOST:ES_PORT"]
                index => "naemongrokfailure"
            }
        }
        else {
            elasticsearch {
                hosts => ["ELASTICSEARCH_HOST:ES_PORT"]
                index => "naemon-%{+YYYY.MM.dd}"
            }
        }
    }
}

Logstash pluging for “perflog”

This Logstash plugin has example of complete configuration for integration with perflog:

input {
  tcp {
    port => 6868  
    host => "0.0.0.0"
    type => "perflogs"
  }
}

filter {
  if [type] == "perflogs" {
    grok {
      break_on_match => "true"
      match => {
        "message" => [
          "DATATYPE::%{WORD:datatype}\tTIMET::%{NUMBER:timestamp}\tHOSTNAME::%{DATA:hostname}\tSERVICEDESC::%{DATA:servicedescription}\tSERVICEPERFDATA::%{DATA:performance}\tSERVICECHECKCOMMAND::.*?HOSTSTATE::%{WORD:hoststate}\tHOSTSTATETYPE::.*?SERVICESTATE::%{WORD:servicestate}\tSERVICESTATETYPE::%{WORD:servicestatetype}",
          "DATATYPE::%{WORD:datatype}\tTIMET::%{NUMBER:timestamp}\tHOSTNAME::%{DATA:hostname}\tHOSTPERFDATA::%{DATA:performance}\tHOSTCHECKCOMMAND::.*?HOSTSTATE::%{WORD:hoststate}\tHOSTSTATETYPE::%{WORD:hoststatetype}"
          ]
        }
      remove_field => [ "message" ]
    }
    kv {
      source => "performance"
      field_split => "\t"
      remove_char_key => "\.\'"
      trim_key => " "
      target => "perf_data"
      remove_field => [ "performance" ]
      allow_duplicate_values => "false"
      transform_key => "lowercase"
    }
    date {
      match => [ "timestamp", "UNIX" ]
      target => "@timestamp"
      remove_field => [ "timestamp" ]
    }
  }
}

output {
  if [type] == "perflogs" {
    elasticsearch {
      hosts => ["127.0.0.1:9200"]
      index => "perflogs-%{+YYYY.MM.dd}"
    }
  }
}

Enabling encryption

Kafka allows you to distribute the load between nodes receiving data and encrypts communication.

Architecture example:

The Kafka installation

Documentation during creation.

Enabling encryption in Kafka

Generate SSL key and certificate for each Kafka broker

keytool -keystore server.keystore.jks -alias localhost -validity {validity} -genkey -keyalg RSA

Configuring Host Name In Certificates

keytool -keystore server.keystore.jks -alias localhost -validity {validity} -genkey -keyalg RSA -ext SAN=DNS:{FQDN}

Verify content of the generated certificate:

keytool -list -v -keystore server.keystore.jks

Creating your own CA

openssl req -new -x509 -keyout ca-key -out ca-cert -days 365
keytool -keystore client.truststore.jks -alias CARoot -import -file ca-cert
keytool -keystore server.truststore.jks -alias CARoot -import -file ca-cert

Signing the certificate

keytool -keystore server.keystore.jks -alias localhost -certreq -file cert-file
openssl x509 -req -CA ca-cert -CAkey ca-key -in cert-file -out cert-signed -days {validity} -CAcreateserial -passin pass:{ca-password}

Import both the certificate of the CA and the signed certificate into the keystore

keytool -keystore server.keystore.jks -alias CARoot -import -file ca-cert
keytool -keystore server.keystore.jks -alias localhost -import -file cert-signed
	

Configuring Kafka Brokers

In /etc/kafka/server.properties file set the following options:

listeners=PLAINTEXT://host.name:port,SSL://host.name:port

ssl.keystore.location=/var/private/ssl/server.keystore.jks
ssl.keystore.password=test1234
ssl.key.password=test1234
ssl.truststore.location=/var/private/ssl/server.truststore.jks
ssl.truststore.password=test1234

and restart the Kafka service

systemctl restart kafka

Configuring Kafka Clients

Logstash

Configure the output section in Logstash based on the following example:

output {
  kafka {
    bootstrap_servers => "host.name:port"
    security_protocol => "SSL"
    ssl_truststore_type => "JKS"
    ssl_truststore_location => "/var/private/ssl/client.truststore.jks"
    ssl_truststore_password => "test1234"
    client_id => "host.name"
    topic_id => "Topic-1"
    codec => json
  }
}

Configure the input section in Logstash based on the following example:

input {
  kafka {
    bootstrap_servers => "host.name:port"
    security_protocol => "SSL"
    ssl_truststore_type => "JKS"
    ssl_truststore_location => "/var/private/ssl/client.truststore.jks"
    ssl_truststore_password => "test1234"
    consumer_threads => 4
    topics => [ "Topic-1" ]
    codec => json
    tags => ["kafka"]
   }
}

Integrations

Setting up Naemon logs

Logstash

1. In op5naemon_beat.conf set up ELASTICSEARCH_HOST, ES_PORT, FILEBEAT_PORT
2. Copy op5naemon_beat.conf to /etc/logstash/conf.d
3. Based on “FILEBEAT_PORT” if firewall is running:

sudo firewall-cmd --zone=public --permanent --add-port=FILEBEAT_PORT/tcp
sudo firewall-cmd --reload

1. Based on amount of data that elasticsearch will receive you can also choose whether you want index creation to be based on moths or days:

index => "op5-naemon-%{+YYYY.MM}"
or
index => "op5-naemon-%{+YYYY.MM.dd}"

1. Copy naemon file to /etc/logstash/patterns and make sure it is readable by logstash process
2. Restart logstash configuration e.g.:

sudo systemct restart logstash

Elasticsearch

1. Connect to Elasticsearch node via SSH and Install index pattern for naemon logs. Note that if you have a default pattern covering settings section you should delete/modify that in naemon_template.sh:

  "settings": {
    "number_of_shards": 5,
    "auto_expand_replicas": "0-1"
  },

1. Install template by running: ./naemon_template.sh

OP5 Monitor

1. On OP5 Monitor host install filebeat (for instance via rpm https://www.elastic.co/downloads/beats/filebeat)

2. In /etc/filebeat/filebeat.yml add:

    #=========================== Filebeat inputs =============================
    filebeat.config.inputs:
      enabled: true
      path: configs/*.yml
   

3. You also will have to configure the output section in filebeat.yml. You should have one logstash output:

    #----------------------------- Logstash output --------------------------------
    output.logstash:
      # The Logstash hosts
      hosts: ["LOGSTASH_IP:FILEBEAT_PORT"]
   

   If you have few logstash instances - Logstash section has to be repeated on every node and hosts: should point to all of them:

    hosts: ["LOGSTASH_IP:FILEBEAT_PORT", "LOGSTASH_IP:FILEBEAT_PORT", "LOGSTASH_IP:FILEBEAT_PORT" ]
   

4. Create /etc/filebeat/configs catalog.

5. Copy naemon_logs.yml to a newly created catalog.

6. Check the newly added configuration and connection to logstash. Location of executable might vary based on os:

    /usr/share/filebeat/bin/filebeat --path.config /etc/filebeat/ test config
    /usr/share/filebeat/bin/filebeat --path.config /etc/filebeat/ test output
   

7. Restart filebeat:

    sudo systemctl restart filebeat # RHEL/CentOS 7
    sudo service filebeat restart # RHEL/CentOS 6
   

Elasticsearch

At this moment there should be a new index on the Elasticsearch node:

curl -XGET '127.0.0.1:9200/_cat/indices?v'

Example output:

	health status index                 uuid                   pri rep docs.count docs.deleted store.size pri.store.size
	green  open   op5-naemon-2018.11    gO8XRsHiTNm63nI_RVCy8w   1   0      23176            0      8.3mb          8.3mb

If the index has been created, in order to browse and visualise the data, “index pattern” needs to be added in Kibana.

Sending OP5 performance data to Elasticsearch node

Below instruction requires that between OP5 node and Elasticsearch node is working Logstash instance.

Elasticsearch

1. First, settings section in op5template.sh should be adjusted, either:

   • there is a default template present on Elasticsearch that already covers shards and replicas then settings sections should be removed from the op5template.sh before executing

   • there is no default template - shards and replicas should be adjusted for you environment (keep in mind replicas can be added later, while changing shards count on existing index requires reindexing it)

       "settings": {
         "number_of_shards": 5,
         "number_of_replicas": 0
       }
     

2. In URL op5perfdata is a name for the template - later it can be search for or modify with it.

3. The “template” is an index pattern. New indices matching it will have the settings and mapping applied automatically (change it if you index name for op5 perfdata is different).

4. Mapping name should match documents type:

    "mappings": {
      "op5perflogs"
   

5. Running op5template.sh will create a template (not index) for OP5 perf data documents.

Logstash

1. The op5perflogs.conf contains example of input/filter/output configuration. It has to be copied to /etc/logstash/conf.d/. Make sure that the logstash has permissions to read the configuration files:

   chmod 664 /etc/logstash/conf.d/op5perflogs.conf
   

2. In the input section comment/uncomment “beats” or “tcp” depending on preference (beats if Filebeat will be used and tcp if NetCat). The port and the type has to be adjusted as well:

    port => PORT_NUMBER
    type => "op5perflogs"
   

3. In a filter section type has to be changed if needed to match the input section and Elasticsearch mapping.

4. In an output section type should match with the rest of a config. host should point to your elasticsearch node. index name should correspond with what has been set in elasticsearch template to allow mapping application. The date for index rotation in its name is recommended and depending on the amount of data expecting to be transferred should be set to daily (+YYYY.MM.dd) or monthly (+YYYY.MM) rotation:

    hosts => ["127.0.0.1:9200"]
    index => "op5-perflogs-%{+YYYY.MM.dd}"
   

5. Port has to be opened on a firewall:

    sudo firewall-cmd --zone=public --permanent --add-port=PORT_NUMBER/tcp
    sudo firewall-cmd --reload
   

6. Logstash has to be reloaded:

    sudo systemctl restart logstash
   

   or

    sudo kill -1 LOGSTASH_PID
   

OP5 Monitor

1. You have to decide wether FileBeat or NetCat will be used. In case of Filebeat - skip to the second step. Otherwise:

   • Comment line:

       54    open(my $logFileHandler, '>>', $hostPerfLogs) or die "Could not open $hostPerfLogs"; #FileBeat
       •	Uncomment lines:
       55 #    open(my $logFileHandler, '>', $hostPerfLogs) or die "Could not open $hostPerfLogs"; #NetCat
       ...
       88 #    my $logstashIP = "LOGSTASH_IP";
       89 #    my $logstashPORT = "LOGSTASH_PORT";
       90 #    if (-e $hostPerfLogs) {
       91 #        my $pid1 = fork();
       92 #        if ($pid1 == 0) {
       93 #            exec("/bin/cat $hostPerfLogs | /usr/bin/nc -w 30 $logstashIP $logstashPORT");
       94 #        }
       95 #    }
     

   • In process-service-perfdata-log.pl and process-host-perfdata-log.pl: change logstash IP and port:

       92 my $logstashIP = "LOGSTASH_IP";
       93 my $logstashPORT = "LOGSTASH_PORT";
     

2. In case of running single op5 node, there is no problem with the setup. In case of a peered environment $do_on_host variable has to be set up and the script process-service-perfdata-log.pl/process-host-perfdata-log.pl has to be propagated on all of OP5 nodes:

    16 $do_on_host = "EXAMPLE_HOSTNAME"; # op5 node name to run the script on
    17 $hostName = hostname; # will read hostname of a node running the script
   

3. Example of command definition (/opt/monitor/etc/checkcommands.cfg) if scripts have been copied to /opt/plugins/custom/:

    # command 'process-service-perfdata-log'
    define command{
        command_name                   process-service-perfdata-log
        command_line                   /opt/plugins/custom/process-service-perfdata-log.pl $TIMET$
        }
    # command 'process-host-perfdata-log'
    define command{
        command_name                   process-host-perfdata-log
        command_line                   /opt/plugins/custom/process-host-perfdata-log.pl $TIMET$
        }
   

4. In /opt/monitor/etc/naemon.cfg service_perfdata_file_processing_command and host_perfdata_file_processing_command has to be changed to run those custom scripts:

    service_perfdata_file_processing_command=process-service-perfdata-log
    host_perfdata_file_processing_command=process-host-perfdata-log
   

5. In addition service_perfdata_file_template and host_perfdata_file_template can be changed to support sending more data to Elasticsearch. For instance, by adding $HOSTGROUPNAMES$ and $SERVICEGROUPNAMES$ macros logs can be separated better (it requires changes to Logstash filter config as well)

6. Restart naemon service:

    sudo systemctl restart naemon # CentOS/RHEL 7.x
    sudo service naemon restart # CentOS/RHEL 6.x
   

7. If FileBeat has been chosen, append below to filebeat.conf (adjust IP and PORT):

    filebeat.inputs:
    - type: log
      enabled: true
      paths:
        - /opt/monitor/var/service_performance.log
        - /opt/monitor/var/host_performance.log
   
    	tags: ["op5perflogs"]
   
   
    output.logstash:
      # The Logstash hosts
      hosts: ["LOGSTASH_IP:LOGSTASH_PORT"]
   

   • Restart FileBeat service:

       sudo systemctl restart filebeat # CentOS/RHEL 7.x
       sudo service filebeat restart # CentOS/RHEL 6.x
     

Kibana

At this moment there should be new index on the Elasticsearch node with performance data documents from OP5 Monitor. Login to an Elasticsearch node and run: curl -XGET '127.0.0.1:9200/_cat/indices?v' Example output:

health status index                      pri rep docs.count docs.deleted store.size pri.store.size
green  open   auth                       5   0          7         6230      1.8mb          1.8mb
green  open   op5-perflogs-2018.09.14    5   0      72109            0     24.7mb         24.7mb

After a while, if there is no new index make sure that:

• Naemon is runnig on OP5 node
• Logstash service is running and there are no errors in: /var/log/logstash/logstash-plain.log
• Elasticsearch service is running an there are no errors in: /var/log/elasticsearch/elasticsearch.log

If the index has been created, in order to browse and visualize the data “index pattern” needs to be added to Kibana.

1. After logging in to Kibana GUI go to Settings tab and add op5-perflogs-* pattern. Chose @timestamp time field and click Create.
2. Performance data logs should be now accessible from Kibana GUI Discovery tab ready to be visualize.

The Grafana instalation

1. To install the Grafana application you should:

   • add necessary repository to operating system:

       [root@logserver-6 ~]# cat /etc/yum.repos.d/grafan.repo
       [grafana]
       name=grafana
       baseurl=https://packagecloud.io/grafana/stable/el/7/$basearch
       repo_gpgcheck=1
       enabled=1
       gpgcheck=1
       gpgkey=https://packagecloud.io/gpg.key https://grafanarel.s3.amazonaws.com/RPM-GPG-KEY-grafana
       sslverify=1
       sslcacert=/etc/pki/tls/certs/ca-bundle.crt
       [root@logserver-6 ~]#
     

   • install the Grafana with following commands:

       [root@logserver-6 ~]# yum search grafana
       Loaded plugins: fastestmirror
       Loading mirror speeds from cached hostfile
        * base: ftp.man.szczecin.pl
        * extras: centos.slaskdatacenter.com
        * updates: centos.slaskdatacenter.com
       =========================================================================================================== N/S matched: grafana ===========================================================================================================
       grafana.x86_64 : Grafana
       pcp-webapp-grafana.noarch : Grafana web application for Performance Co-Pilot (PCP)
     
         Name and summary matches only, use "search all" for everything.
     
       [root@logserver-6 ~]# yum install grafana
     

   • to run application use following commands:

       [root@logserver-6 ~]# systemctl enable grafana-server
       Created symlink from /etc/systemd/system/multi-user.target.wants/grafana-server.service to /usr/lib/systemd/system/grafana-server.service.
       [root@logserver-6 ~]#
       [root@logserver-6 ~]# systemctl start grafana-server
       [root@logserver-6 ~]# systemctl status grafana-server
       ● grafana-server.service - Grafana instance
          Loaded: loaded (/usr/lib/systemd/system/grafana-server.service; enabled; vendor preset: disabled)
          Active: active (running) since Thu 2018-10-18 10:41:48 CEST; 5s ago
            Docs: http://docs.grafana.org
        Main PID: 1757 (grafana-server)
          CGroup: /system.slice/grafana-server.service
                  └─1757 /usr/sbin/grafana-server --config=/etc/grafana/grafana.ini --pidfile=/var/run/grafana/grafana-server.pid cfg:default.paths.logs=/var/log/grafana cfg:default.paths.data=/var/lib/grafana cfg:default.paths.plugins=/var...
     
        [root@logserver-6 ~]#
     

2. To connect the Grafana application you should:

   • define the default login/password (line 151;154 in config file)

       [root@logserver-6 ~]# cat /etc/grafana/grafana.ini
       .....
       148 #################################### Security ####################################
       149 [security]
       150 # default admin user, created on startup
       151 admin_user = admin
       152
       153 # default admin password, can be changed before first start of grafana,  or in profile settings
       154 admin_password = admin
       155
     

….. - restart grafana-server service:

		[root@logserver-6 ~]# systemctl restart grafana-server

- Login to Grafana user interface using web browser: *http://ip:3000*

- use login and password that you set in the config file.

1. Use below example to set conection to Elasticsearch server:

   

The Beats configuration

Kibana API

Reference link: https://www.elastic.co/guide/en/kibana/master/api.html

After installing any of beats package you can use ready to use dashboard related to this beat package. For instance dashboard and index pattern are available in /usr/share/filebeat/kibana/6/ directory on Linux.

Before uploading index-pattern or dashboard you have to authorize yourself:

1. Set up login/password/kibana_ip variables, e.g.:

    login=logserver
    password=my_password
    kibana_ip=10.4.11.243
   

2. Execute command which will save authorization cookie:

    curl -c authorization.txt -XPOST -k "https://${kibana_ip}:5601/login" -d "username=${username}&password=${password}&version=6.2.3&location=https%3A%2F%2F${kibana_ip}%3A5601%2Flogin"
   

3. Upload index-pattern and dashboard to Kibana, e.g.:

   curl -b authorization.txt -XPOST -k "https://${kibana_ip}:5601/api/kibana/dashboards/import" -H 'kbn-xsrf: true' -H 'Content-Type: application/json' -d@/usr/share/filebeat/kibana/6/index-pattern/filebeat.json
   curl -b authorization.txt -XPOST -k "https://${kibana_ip}:5601/api/kibana/dashboards/import" -H 'kbn-xsrf: true' -H 'Content-Type: application/json' -d@/usr/share/filebeat/kibana/6/dashboard/Filebeat-mysql.json
   

4. When you want to upload beats index template to Ealsticsearch you have to recover it first (usually you do not send logs directly to Es rather than to Logstash first):

   /usr/bin/filebeat export template --es.version 6.2.3 >> /path/to/beats_template.json
   

5. After that you can upload it as any other template (Access Es node with SSH):

   curl -XPUT "localhost:9200/_template/op5perfdata" -H'Content-Type: application/json' -d@beats_template.json
   

Wazuh integration

OP5 Log Analytics can integrate with the Wazuh, which is lightweight agent is designed to perform a number of tasks with the objective of detecting threats and, when necessary, trigger automatic responses. The agent core capabilities are:

• Log and events data collection
• File and registry keys integrity monitoring
• Inventory of running processes and installed applications
• Monitoring of open ports and network configuration
• Detection of rootkits or malware artifacts
• Configuration assessment and policy monitoring
• Execution of active responses

The Wazuh agents run on many different platforms, including Windows, Linux, Mac OS X, AIX, Solaris and HP-UX. They can be configured and managed from the Wazuh server.

Deploying Wazuh Server

https://documentation.wazuh.com/current/installation-guide/installing-wazuh-server/index.html#

Deploing Wazuh Agent

https://documentation.wazuh.com/current/installation-guide/installing-wazuh-agent/index.html

Filebeat configuration

BRO integration

Troubleshooting

Recovery default base indexes

If you lost or damage following index:

    |Index name		 | Index ID              |
    |----------------|-----------------------|
    | .security      |Pfq6nNXOSSmGhqd2fcxFNg |
    | .taskmanagement|E2Pwp4xxTkSc0gDhsE-vvQ |
    | alert_status   |fkqks4J1QnuqiqYmOFLpsQ |
    | audit          |cSQkDUdiSACo9WlTpc1zrw |
    | alert_error    |9jGh2ZNDRumU0NsB3jtDhA |
    | alert_past     |1UyTN1CPTpqm8eDgG9AYnw |
    | .trustedhost   |AKKfcpsATj6M4B_4VD5vIA |
    | .kibana        |cmN5W7ovQpW5kfaQ1xqf2g |
    | .scheduler_job |9G6EEX9CSEWYfoekNcOEMQ |
    | .authconfig    |2M01Phg2T-q-rEb2rbfoVg |
    | .auth          |ypPGuDrFRu-_ep-iYkgepQ |
    | .reportscheduler|mGroDs-bQyaucfY3-smDpg |
    | .authuser      |zXotLpfeRnuzOYkTJpsTaw |
    | alert_silence  |ARTo7ZwdRL67Khw_HAIkmw |
    | .elastfilter   |TtpZrPnrRGWQlWGkTOETzw |
    | alert          |RE6EM4FfR2WTn-JsZIvm5Q |
    | .alertrules    |SzV22qrORHyY9E4kGPQOtg |

You may to recover it from default installation folder with following steps:

1. Stop Logstash instances which load data into cluster

    systemctl stop logstash
   

2. Disable shard allocation

    PUT _cluster/settings
    {
      "persistent": {
        "cluster.routing.allocation.enable": "none"
      }
    }
   

3. Stop indexing and perform a synced flush

    POST _flush/synced
   

4. Shutdown all nodes:

    systemctl stop elasticsearch.service
   

5. Copy appropriate index folder from installation folder to Elasticsearch cluster data node folder (example of .auth folder)

    cp -rf ypPGuDrFRu-_ep-iYkgepQ /var/lib/elasticsearch/nodes/0/indices/
   

6. Set appropriate permission

    chown -R elasticsearch:elasticsearch /var/lib/elasticsearch/
   

7. Start all Elasticsearch instance

    systemctl start elasticsearch
   

8. Wait for yellow state of Elasticsearch cluster and then enable shard allocation

    PUT _cluster/settings
    {
      "persistent": {
        "cluster.routing.allocation.enable": "all"
      }
    }
   

9. Wait for green state of Elasticsearch cluster and then start the Logstash instances

    systemctl start logstash
   

Upgrades

Manual update of OP5 from version 6.1.1 to 6.1.2

The OP5 Log Anlytics 6.1.1 update should be done by copying new versions of files to the appropriate directories. The source installation directory is: /root/pkg_6.1.2*

Data node update

1. Go to installation directory that contain OP5 Log Analytics 6.1.2 files:

    cd /root/pkg_6.1.2
   

2. Stop the Elasticsearch service

    systemctl stop elasticsearch
   

3. Backup the elasticsearch-auth plugin

    cp –rf /usr/share/elasticsearch/plugins/elasticsearch-auth /root/backup/
   

4. Copy the new version of the plugin from the installation directory:

    cp -rf elastisearch/elasticearch-auth /usr/share/elasticsearch/plugins/
   

5. Grant the appropriate permissions for directories:

    chown -R elasticseach:elasticsearch /usr/share/elasticsearch
   

6. Start the Elasticsearch service

    systemctl start elasticsearch
   

Client Node update

1. Check if the following RPM packages have been installed:

    yum install fontconfig freetype freetype-devel fontconfig-devel libstdc++ urw-fonts net-tools ImageMagick ghostscript poppler-utils
   

2. Stop the Kibana and Alert services:

    systemctl stop kibana
    systemctl stop alert
   

3. Delete the contents of the directory /usr/share/kibana/optimize/bundles/

    rm -rf /usr/share/kibana/optimize/bundles/*
   

4. Delete the contents of the directory /usr/share/kibana/plugins

    rm -rf /usr/share/kibana/plugins/*
   

5. Backup current Alert rules folder

    cp -pr /opt/alert/rules /root/backup/alert/rules
   

6. Delete old version of Alert plugin

    rm -rf /opt/alert
   

7. Delete old version of AI plugin

    rm -rf /opt/ai
   

8. Copy the Kibana plugins from the installation directory

    cp -rf kibana/plugins/* /usr/share/kibana/plugins/
   

9. Copy the Alert plugin from the installation directory

    /bin/cp -rf alert /opt/alert
   

10. Copy Alert rules folder

     cp -pr /root/backup/alert/rules/* /opt/alert/rules/
    

11. Copy the Alert plugin from the installation directory

      /bin/cp -rf ai /opt/ai
    

12. Unpack the node.js modules

     tar -xf /usr/share/kibana/plugins/node_modules.tar -C /usr/share/kibana/plugins/
    

13. Delete the unnecessary tar.gz archive

     /bin/rm -rf /usr/share/kibana/plugins/node_modules.tar
    

14. Give the right permissions to the Kibana directory

     chown -R kibana:kibana /usr/share/kibana
    

15. Perform the start of Kibana and Alert

     systemctl start kibana
     systemctl start alert
    

Upgrade the OP5 Log Analytics with RPM packets

Data node (every Elasticsearch installation) update

1. Currently updateing with RPMs is not supported - refer to manual update section when upgrading data node.

Client Node update (GUI)

1. Stop the Kibana and Alert services:

    systemctl stop kibana
    systemctl stop alert
   

2. Delete bundles /usr/share/kibana/optimize/bundles/

    rm -rf /usr/share/kibana/optimize/bundles/*
   

3. Delete the contents of the directory /usr/share/kibana/plugins

    rm -rf /usr/share/kibana/plugins/*
   

4. Run update with following command:

    rpm -Uvh --replacefiles op5-log-analytics-client-node-6.1.2-1.x86_64.rpm
   

5. Give the right permissions to the Kibana directory

    chown -R kibana:kibana /usr/share/kibana
   

6. Perform the start of Kibana and Alert

    systemctl start kibana
    systemctl start alert
   

The Agents module.

The Agents module is used for the central management of agents used in OP5 Log Analytics such as Filebeat, Winlogbeat, Packetbeat, Metricbeat.# Agent installation # All necessary components can be found in the installation folder ${installation_folder}/utils/agents_bin.

Component modules

The software consists of two modules:

• Plugin Agents - installation just like any standard Kibana plugin. Before you run the module for the first time, you must add the mapping for the .agents index with the create_temlate.sh script
• MasterAgent software - installed on host with agent (like beats);

Table of configuration parameter for Agent software

	|Parameter            |Work type             |Required |Defult value           |Description                                 |
	|---------------------|----------------------|---------|-----------------------|--------------------------------------------|
	|port                 |Agent                 |No       |40000                  |The port on which |the agent is listening   |
	|host                 |Agent                 |No       |Read from system       |The address on which the agent is listening |
	|hostname             |Agent                 |No       |Read from system       |Host name (hostname)                        |
	|autoregister         |Agent                 |No       |24                     |How often the agent's self-registration should take place. Time in hours |
	|metricbeat_path      |Agent                 |No       |./                     |Catalog for meatricbeat                     |
	|filebeat_path        |Agent                 |No       |./                     |Directory for filebeat                      |
	|winlogbeat_path      |Agent                 |No       |./                     |Catalog for winlogbeat                      | 
	|packetbeat_path      |Agent                 |No       |./                     |Catalog for packetbeat                      |
	|custom_list          |Agent                 |No       |Not defiend            |List of files and directories to scan. If a directory is specified, files with the yml extension are registered with it. The file / directory separator is the character ";" |
	|createfile_folder    |Agent                 |No       |Not defiend            |List of directories where files can be created. The catalogs are separated by the symbol ";". These directories are not scanned for file registration.
	|logstash             |Agent                 |No       |https://localhost:8080 |Logstash address for agents                  |
	|https_keystore       |Agent and Masteragent |No       |./lig.keystore         |Path to the SSL certificate file.            |
	|https_keystore_pass  |Agent and Masteragent |No       |admin                  |The password for the certificate file        |
	|connection_timeout   |Agent and Masteragent |No       |5                      |Timeout for https calls given in seconds.    |
	|connection_reconnect |Agent and Masteragent |No       |5                      |Time in seconds that the agent should try to connect to the Logstash if error occur |

Installing the agent software

The Agent’s software requires the correct installation of a Java Runtime Environment. The software has been tested on Oracle Java 8. It is recommended to run the Agent as a service in a given operating system.

1. Generating the certificate

   The Logshash, Agent and Masteragent use the same certificate file. To generate a file, use the command:

    	keytool -genkey -alias aka -keypass simulator -keystore lig.keystore -storepass simulator
   

2. Logstash configuration

   • input

       input {
         http {
           ssl => true
           keystore => "/opt/lig.keystore"
           keystore_password => "simulator"
           tags => ["agents"]
         }
       }
     

   • output

       output {
         if "agents" in [tags] {
       elasticsearch {
           hosts => "localhost:9200"
           manage_template => false
           index => ".agents" 
           document_type => "doc" 
         }
         }
       }
     

3. Linux host configuration

   • Download MasterBeatAgent.jar and agent.conf files to any desired location;
   • Upload a file with certificates generated by the keytool tool to any desired location;
   • Update entries in the agent.conf file (the path to the key file, paths to files and directories to be managed, the Logstash address, etc.);
   • The agent should always be run with an indication of the working directory in which the agent.conf file is located;
   • The Agent is started by the java -jar MasterBeatAgent.jar command.

4. Windows host configuration

   • Installation in Windows is analogous to installing on a Linux system;
   • To install the agent as a service, you can use the wrapper ${installation_folder}/utils/agents_bin/wrapper;
   • The sample of agents.exe and agents.xml files are in the agents_bin/wrapper directory;
   • As a working directory, set the directory where the agent configuration file is located.

The agent management

The GUI console is used to manage agents. In the Agetns tab, you can find a list of connected agents. There are typical information about agents such as:

• Host name;
• OS name;
• IP Address;
• TCP port;
• Last revision;

Additionally, for each connected agent, you can find action buttons such as:

• Drop - to remove the agent configuration from the GUI;
• Create - to create new configuration files;
• Show - it is used to display the list of created configuration files;

Creating a new configuration file

To add a new configuration file press the Create button, add a new file name, add a new path where the file should be saved and the context of the new configuration file. The new file will be saved with the extension * .yml.

Editing configuration file

To display a list of configuration files available for a given host, press the Show button.

A list of configuration files will be displayed, and the following options for each of them:

• Show - displays the contents of the file;
• Edit - edit the contents of the file;
• Delete - deletes the file.

To edit the file, select the Edit button, then enter the changes in the content window, after finishing select the Submit button.

Monitoring

Skimmer

OP5 Log Analytics uses a monitoring module called Skimmer to monitor the performance of its hosts. Metrics and conditions of services are retrieved using the API.

The services that are supported are:

• Elasticsearch;
• Logstash;
• Kibana;
• Metricbeat;
• Pacemaker;
• Zabbix;
• Zookeeper;
• Kafka;
• Httpbeat;
• Elastalert;
• Filebeat

and other.

Skimmer Installation

The RPM package skimmer-x86_64.rpm is delivered with the system installer in the “utils” directory:

	cd $install_direcorty/utils
	yum install skimmer-x86_64.rpm -y

Skimmer service configuration

The Skimmer configuration is located in the /usr/share/skimmer/skimmer.conf file.

# index name in elasticsearch
index_name = skimmer
index_daily = true
# type in elasticsearch index
index_type = _doc
# user and password to elasticsearch api
elasticsearch_auth = logserver:logserver
# available outputs
elasticsearch_address = 127.0.0.1:9200
# logstash_address = 127.0.0.1:6110
# retrieve from api
elasticsearch_api = 127.0.0.1:9200
# logstash_api = 127.0.0.1:9600
# path to log file
log_file = /tmp/skimmer.log
# daemonize
daemonize = true
# comma separated OS statistics selected from the list [zombie,vm,fs,swap,net,cpu]
os_stats = zombie,vm,fs,swap,net,cpu
# comma separated process names to print their pid
processes = /usr/sbin/sshd,/usr/sbin/rsyslogd
# comma separated systemd services to print their status
systemd_services = elasticsearch,logstash,kibana
# comma separated port numbers to print if address is in use
port_numbers = 9200,9300,9600,5601
# path to directory containing files needed to be csv validated
csv_path = /tmp/csv_dir

After the changes in the configuration file, restart the service.

systemctl restart skimmer

Skimmer GUI configuration

To view the collected data by the skimmer in the GUI, you need to add an index pattern.

Go to the “Management” -> “Index Patterns” tab and press the “Create Index Pattern” button. In the “Index Name” field, enter the formula skimmer- *, and select the “Next step” button. In the “Time Filter” field, select @timestamp and then press “Create index pattern”

In the “Discovery” tab, select the skimmer- * index from the list of indexes. A list of collected documents with statistics and statuses will be displayed.

CHANGELOG

Version 6.1.6

Added

• BREAKING CHANGE: Support of simple upgrade procedure alert indices have to be reindexed
• Alerting module upgraded
• System indices created automaticly durring install
• Improved settings for system indices (priority, shard count, automatic replicas)
• Validate playbooks button when updating alert rule
• Order of plugins is no longer random
• Reports plugin now takes roles into consideration when creating and browsing generated reports
• Object permission lists are now sorted
• Improved CSV export field list (sorting and bigger size)
• DevTools enabled/disabled directive added to default kibana.yml
• Timelion enabled/disabled directive added to default kibana.yml

CHANGED

• bugfix: CVE-2019-7608
• bugfix: CVE-2019-7609
• bugfix: CVE-2018-3830
• bugfix: filtering logo extension during upload and report generation
• bugfix: improved verification for Create User
• bugfix: report scheduling for AD users
• bugfix: downloading jpeg exports now returns correct response header
• bugfix: could not set risk category to zero
• bugfix: IE11 compability fix when creating new alert
• bugfix: Admin users see all alerts
• bugfix: Error message if you try create new alert but it already exists

Version 6.1.5

• BREAKING CHANGE: audit index is from now on created with type “doc” and date field “@timestamp”. Old index is not compatible and should be deleted before update:
• Turn of audit logging. In Kibana -> Settings and unmark all in “Update Audit Setting” section.
  • Delete the audit index
  • Update elasticsearch-auth
  • Turn on audit logging.

• Risk Management for Alerts - User can create custo categories for field attributes like Hostname, Hostip, Username. Once the alert is triggred, the result get score amplification calculated from object categories.
• Alert rule importance - introduction of new value for each alerts that is correlated with objedct category and helps identify
• When creating alerts now we have the ability Test the rule before scheduling this
• Playbook introduction - ability to create simple editible instructions(+scripts) that system oerator should follow when Alert is triggered
• Verify IP on blacklists - if the Alerrt is triggred for IP, Verify button let us check its reputaion
• When creating alerts operatos get ability to validate the alert and find most suitable playbook for it. Playbook list is automaticly sorted.
• User get email notification when Incident is attached to them. New email field in user tab.
• IP’s are correlated towards Bad IP reputation list
• Introduction of Incidents. Alerts are now turned into Incidents, with assigned operator and its status
• Regular user can configure own Alerts
• Netflow, jflow, sflow support
• Provided interface for running custom, external, AI jobs created in own programming language

• Audit index is from now created with type “doc” and date field “@timestamp”
• Better Radius authentication supoort
• System auditing corrections

CHANGED

• bugfix: in intelligence module api
• bugfix: fixes in sorting alerts

Version 6.1.3

Added

• Securing all the endpoints of elasticsearch APIs
• New configuration option: elastfilter.proxytimeout
• Cleaning unnecessary objects in kibana indices
• Upgrade default logstash to 6.6.2
• Mobile App for Energy Logserver that works with : Log Analytics, Energy Logsrver, pure ELK. x-pack is extra paid. Available for Android and ios. https://play.google.com/store/apps/details?id=com.logserver.mobile

CHANGED

• bugfix: problem with creating scheduled reports
• bugfix: SSO login not working due to more secure java.policy
• bugfix: Performance issue while using non admin account
• bugfix: Java exception while useing elasticsearch-plugin (ES_JAVA_OPTS moved to jvm.options)
• bugfix: default encoding for es2csv changed to utf-8 (csv export with polish characters)

Version 6.1.2

Added

• Intelligence API
• Kibana API update
• Caching for index list and roles for user to handle the high CPU usage on master node
• Export task as HTML
• Dashboard report as JPEG
• Additional logging in debug mode in elasticsearch-auth plugin
• GC1 used as default Garbage Collector
• NioFS as default Store Type
• Compression for http & transport enabled
• Product Version tab in Config module
• New Agents feature for central beats/agents management

CHANGED

• bugfix: user session timeouts
• bugfix: problem with reports generation using 5601->443 port redirection
• bugfix: problem with removing a large number of objects from Kibana
• bugfix: timepicker on export to csv reports
• bugfix: special chars in passwords
• bugfix: java.policy - binding elasticsearch to 0.0.0.0
• bugfix: service_principal_name - is no longer required directive when configuring work with AD/LDAP

Version 6.1.1

Added

• Default template with compression only [elasticsearch]
• Secured LDAP/AD password in configuration files [elasticsearch]

CHANGED

• bugfix: filter config - linux-geoip [logstash]
• bugfix: intelligence template

Version 6.1.0

• Upgrade core to 6.2.4 [elasticsearch,kibana,logstash]
• Support for all beats agents in filters and dashboards
• Providing default Audit and Alert dashboard
• Change in Intelligence Spark data provide - 1:20 speed improvement
• Intelligence not sensitive on data types
• Better Intelligence preview capabilities·
• Intelligence Count & Trend improvement
• Technology specific dashboards : Windows, Linux, Network
• Technology specific alerts : Windows, Linux, Network
• OP5 Monitor perf data support with filtering and dashboard
• UTF-8 support in custom PDF reports

CHANGED

• bugfix: logo/title/comment in reports module now as optional
• bugfix: java.policy·
• bugfix: Alert Status in Alert module
• bugfix: Percentagematch and Metricaggregation rules fix in Alert module
• bugfix: Deleting Alert rule cause Alert Disable

Version 6.0.2

Added

• SSO onboarded to 6.x stack
• Custom Logo on PDF Reports including title and comment
• Data Table Head - new visualization type·
• Controls Plus - new vizualization type·
• “Count in Time” as type in Intelligence module
• Nasted.fields support in Intelligence module
• GUI for Scheduler module
• support for all beats agents in filters and dashboards
• providing default audit dashboard

CHANGED

• bugfix: Removed ‘:’ from escaped characters - “Boo” message
• bugfix: Missing directories for reports
• bugfix: Removed unessesary files from deps

Version 6.0.1

Added

• Functional indexess with dots .kibana, .security, .auth
• Login module onboarded to 6.x stack
• License module onboarded to 6.x stack
• Default roles: alert, intelligence, kibana·
• Export to CSV [Task Management] onboarded to 6.x stack
• Export do PDF [Reports] onboarded to 6.x stack
• PDF Scheduler onboarded to 6.x stack
• AD integrations onboarded to 6.x stack·