IBMStreams/streamsx.demo.logwatch

Name: streamsx.demo.logwatch

Owner: IBM Streams

Description: (Incubation) Demonstration SPL application for detecting system breakins.

Created: 2014-05-16 17:38:27.0

Updated: 2017-10-12 19:34:21.0

Pushed: 2017-10-12 19:34:20.0

Homepage:

Size: 143

Language: Makefile

GitHub Committers

UserMost Recent Commit# Commits

Other Committers

UserEmailMost Recent Commit# Commits

README

streamsx.demo.logwatch

The purpose of the LogWatch family of applications is to teach concepts in SPL and Streams through a self-contained, small application with a well-defined problem statement. The application in the language sub-namespace is a language tutorial, and the application in the performance sub-namespace builds on that starting point, but with the goal of demonstrating strategies to improve performance.

Problem Statement

The LogWatch family of applications are in the security domain. They watch the system messages file on a Linux system, (usually found at /var/log/messages), flagging security breaches. Breakins are detected using the following observation:

If the same remote host attempts to login many times in a short period of time, then succeeds, it is likely a breakin.

We can also track long and medium term suspicious behavior, based on the number of failed login attempts in a time frame from the same remote host. While applications in the LogWatch family be be implemented in different ways, they all solve the above problem.

Iterative developement:

In all of the applications, SPL files in the etc directory show iterative development of that particular application. Since these applications are intended for teaching purposes, the final program is not the primary goal. Instead, the goal is to teach a development process. The sequence of SPL programs in the etc directory for each application demonstrate the iterative process used to accomplish the end program.

Contact

Scott Schneider, scott.a.s@us.ibm.com


This work is supported by the National Institutes of Health's National Center for Advancing Translational Sciences, Grant Number U24TR002306. This work is solely the responsibility of the creators and does not necessarily represent the official views of the National Institutes of Health.