Showing posts with label UEBA. Show all posts
Showing posts with label UEBA. Show all posts

Tuesday, December 3, 2019

Foundation of UEBA

According to Gartner there are three dimensions we should think of when talking about UEBA. They are
1. Data Source
UEBA implementation should be as such that it ingest data from a data repository. Data repository could be data warehouse or SIEM.

2. Use Cases
UEBA should provide an information about an anomalies of user and other entities. It needs to continuously monitor, detect and alert those anomalies. So to make it more effective UEBA should support multiple use cases.

3. Analytics
UEBA needs to implement various types of approaches like Machine learning, Statistical model, threat signatures, threat rules etc. to detect an anomalies.

Thursday, September 5, 2019

UEBA Use Cases

With integration of UEBA in SIEM we wish to solve various cases. Some of the use cases that would particularly be of interest to the system administrator and security analyst can be listed as below

1. Account Misuse or Malicious Insider
Account misuse or Malicious Insider is a condition when employee of an organization with certain level of privileged access tends to perform attack against an organization. It is quite difficult to detect such situations only looking at the log files. Also, regularly monitoring security events will not solve the problem. Insider is someone who know what is where and how it can be accessed. Moreover, insider also happens to know workaround to go undetected.

We can, however, detect such malicious intent of an insider by defining a baseline of who access what, when and/or how. Here, UEBA helps by baselining user's behavior and detecting when things do not confirm to normal behavior.


2. Compromised Accounts
Once attacker infiltrate into an organization, they compromise privilege account or they try to escalate the privilege of the account they infiltrated with. It is because only privilege level account can be used to access sensitive and confidential information within an organization. During the process of escalation of account or actual compromise events various audit log gets recorded. UEBA detects various bad activities from compromised account.

3. Internal Reconnaissance
With Internal reconnaissance attacker collects internal information about a target network. This allows attacker to effectively move through the network and conduct further activities. Attacker uses several methods including hosts, networks, and processes enumeration which lists all relevant information to develop an understanding of where an attacker has landed on a network, and to plan their next move. UEBA can easily detect internal reconnaissance using various machine learning techniques.

4. Lateral Movement
Once attacker infiltrate into an organization they then begins to look for the sensitive information. During this process they move laterally through a network as they search for the key data and assets that are ultimately the target of their attack campaigns. Without UEBA, lateral movement is very difficult to detect.

5. Data Exfiltration
Unauthorized transfers of data, manually or automatically via malicious programs, over a network is known as data exfiltration. This is one of the serious concern in today's business. Leak of business critical information would cost a lot to that organization. This use case always remains the most 'looked for' use cases in business. However, to ascertain whether or not exfiltration has happened we need to trace long history of sequence of events. Without UEBA this will be very difficult.

6. Incident Prioritization
Tons of incidents are generated every hour and it is quite impossible to correctly handle all of such incidents. In this scenario we require something that could prioritize incidents according to their severity, thus allowing system administrator to properly handle each notification in turn.

UEBA needs to rank the incidents as per their priority so that security professionals can focus on the most likely threats to their critical business assets.

Wednesday, September 4, 2019

UEBA and SIEM

User and Entity Behavior Analytics (UEBA) as name suggests provides behavioral analytics of user and other entities using different learning algorithms. It is vital for UEBA to get data from different log sources to perform its analysis and provide result.

Security Incident and Event Management (SIEM) in essence collects, stores and monitors logs from different devices.

This means UEBA and SIEM are closely related. One collects and stores logs while other require centrally collected logs to provide output. Also, if you look into Gartner's vision of next-generation SIEM solution, UEBA should be the built-in functionality in SIEM. Advanced analytics and user monitoring (baselining and advanced analytics to analyze access and authentication data, establish user context and report on suspicious behavior.) are two fundamental capabilities SIEM should posses. Thus, UEBA and SIEM goes hand-in-hand.

Saturday, January 19, 2019

Threat Hunting - What you can Discover

Various types of threats exists in our network and we need to be able to discover them before any harm is done. However, not all threats are easily detectable and needs behavior analysis. User and entity behavior analysis (UEBA) is one of the methods which is employed to detect threats using machine learning techniques. Hunting for a threat requires going through each steps in cyber kill chain. Following picture gives an overview of what cyber kill chain is composed of


After successfully implementing cyber kill chain in your infrastructure following are the threats you can discover through hunting process...


Happy hunting!!!