A different take: SIEM

Saurabh Sharma

Security space is an ever evolving. It never stays stagnant; everyday you see a new exploit, a new hack. As an individual, you may have Antivirus installed on your system, Firewall – UP, ports blocked yet you are not free from threat

If you have….

  1. Digital footprint: Use any of the social apps e.g. Facebook, Twitter etc.
  2. Use an email: Gmail, yahoo, live etc.
  3. Enter passwords (even though super secure)
  4. Shop online,
  5. Heck even if you have a smart phone

you are at risk, everyday.

Even if you do not have or use any of the 5 above, you still have your identity, right? SSN, or UID?

Your Biometrics fingerprints, eyeprints, stored in a DB govt or otherwise.

Just think about it; if this is for an individual what about an organization?

Organizations & SIEM

With the ever evolving space of security the organization are also upping their game. You have dedicated analysts (individually or part of SoC or NoC) analyzing information using variety of tools, one such tool is Splunk (very prevalent), Elastic etc.

Perspective

What is threat?

A declaration of an intention or determination to inflict punishment, injury, etc., in retaliation for, or conditionally upon, some action or course; menace:

Threat is always contextual, specifically in the space of security e.g.

[Error: 401] Unauthorized A single authentication failure where say John Doe, operating from Australia tried to access his account in his organization and misspelled his password resulting an error.

Is this threat?

It might be a a genuine mistake or lets add some spices to the situation

The request came in at a time lets say 5:30 in the evening (AEST) but came in from an IP that is from North Korea

Rings a bell?

So now we are thinking

Volume Metrics

Too many unauthorized errors coming might be an anomaly, but may or may not be threat.

E.g. A configured password was just changed and the dev guy forgot to change one location and hence it may be false positive.

Behavior Metrics

Behavior is a pattern that is observed over a time for an entity. This entity can be a user or a resource.

John regularly logs in from Sydney around 8:00 to 8:30 AM and logs out by 6:00 to 6:30 PM sometimes he logs in on weekends, but not regularly.

Param : One

Today, John logs in at 7:00 PM on Friday and from Japan – Suspicious?

This is the deviations from historical pattern, which defines John Doe, as a regular user logging in at 8:00 AM logging out 6:00 PM everyday

Param : Two

The last login location was Australia, and within 15 minutes John logs in from Japan? Suspicious again?

This is what is called anomaly from the regular behavior and may or may not be a threat, considering John may be on a regular visit to Japan, but what is suspicious is he was logged in from Australia 15 minutes ago so physically it is impossible to make a quantum leap in 15 minutes from one geographic location to another.

So this definitely looks like a genuine threat.

Spatial: Metric

Spatial is the location element of an entity, which impacts a Volume or Behavior metric in a way. It is not just an attribute of an entity it can enrich the information in a decisive way.

SIEM: Security Information & Event Management

SIEM is a process, approach or tools to monitor, analyze, evaluate and identify threats in an organization.

It collects security data from appliances like servers, firewalls, domain controllers, and more. It stores & normalizes data to discover trends, threats, and enable organizations to plug the gaps.

The heart of all analysis is the data which is refined, correlated, enriched if required and normalized.
  • Identify high-risk threats with near real-time correlation and behavioral anomaly detection.
  • Detect vulnerabilities, manage risks and identify high-priority incidents among billions of data points.
  • Gain full visibility into network, application and user activity.

Tools

I have used mostly Splunk and Elastic, but the solutions are plenty depending on your pocket and size of the organization

Helpful links