Characteristics of Internet Background Radiation

1
Characteristics of Internet Background Radiation

• Authors
• Ruoming Pang,
• Vinod Yegneswaran,
• Paul Barford,
• Vern Paxson,
• Larry Peterson
• Publisher
• ACM Internet Measurement Conference
• (IMC), 2004
• Presented by
• Chowdhury, Abu Rahat

2
Todays Outline

• The Authors and their Problem Statements
• Objective Terminology
• The study and Network Telescope
• Measurement Methodology
• Passive Measurement
• Active Measurement
• Comments.

3
The Authors
Vern Paxson Associate Professor EECS Department
of UC Berkeley,
Vinod YegneswaranGrad Student Computer Science
and Statistics University of Wisconsin
Ruoming Pang Software engineer Google NY
Paul Barford Assistant Professor, Department of
Computer SciencesUniversity of Wisconsin-Madison
Larry L. Peterson Professor Department of
Computer Science Princeton, NJ 08544
Current Research Projects and ThrustsMeasurement,
analysis, and security of wide area networked
systems and network protocols
4
The Problem

• Background radiation reflects fundamentally
  nonproductive traffic, either malicious or
  benign. While the general presence of background
  radiation is well known to the network operator
  community, its nature has yet to be broadly
  characterized
• Goals of Characterization
• What is all this nonproductive traffic trying to
  do?
• How can we filter it out to detect new types of
  malicious activity?

5
Outline

• The Authors and their Problem Statements
• Objective Terminology
• The study and Network Telescope
• Measurement Methodology
• Passive Measurement
• Active Measurement
• Comments

6
Objective

• To characterize Background Radiation based on
• Types of attack, behavior, traffic composition,
  frequency, target networks, etc.
• Secondary objectives
• Development of an effective traffic filtering
  system
• Use of active responders to effectively identify
  the objective of attacks

7
Natural Background Radiation

• We are all exposed to ionizing radiation from
  natural sources at all times. This radiation is
  called natural background radiation, and its main
  sources are the following
• Radioactive substances in the earth's crust
• Emanation of radioactive gas from the earth
• Cosmic rays from outer space which bombard the
  earth

Source Google Earth
8
Internet Background Radiation

• The Baseline Noise of Internet traffic
• Every IP address---even an unused
  one---receives packets constantlySo
  Fundamentally nonproductive traffic.
• Traffic sent to unused addresses.
• Nonproductive traffic malicious
• (flooding backscatter, hostile scan, spam)
• OR benign (misconfigurations).
• Pervasive nature (hence background).

9
Backscatter
Source MVS01
10
Background Radiation

• The volume of this traffic is not minor. For
  example, traffic logs from LBL for an
  arbitrarily-chosen day show that a total of about
  8 million connection attempts (2/3 of the total)

Background Radiation
Benign
Malicious
Misconfiguration
Backscatters
Worms
Scan for Vulnerability
11
The Study

• Why do we study it?
• To understand Internet malware in action
• This paper is the first broad characterization
  of Internet background radiation
• Focus traffic semantics
• What is the traffic trying to do at application
  level?
• Measurement methodology
• How to extract the meaning of background
  radiation ?

12
Measurement ApparatusNetwork Telescope
Unused but globally reachable IP Addresses
Their main telescopes Lawrence Berkeley
National Lab Size 1,280 addresses
13
Outline

• The Authors and their Problem Statements
• Objective Terminology
• The study and Network Telescope
• Measurement Methodology
• Passive Measurement
• Active Measurement
• Comments.

14
Measurement Methodology PassiveHit Pattern
What is the type and volume of observed traffic
without actively responding to any packet?
15
How Often Do We See a Packet?

• Feb 2006 at Lawrence Berkeley Lab
• (Average on 1,280 IPs over period of a week)
• 342 packets / destination IP / day
• gt A packet every 4 minutes on any IP
• But, how are radiation packets distributed
• Among destination IPs? (Hotspot?)
• Over time

Source Ruoming Pang
16
Distribution over Destination IPs
Number of packets per destination IP received
over a week
17
Distribution over Destination IPs
Packets are in general evenly distributed among
destinations The biggest hotspot receives lt 1
of packets
18
Number of Source IPs Per Hour
Number of source IPs also vary over time But not
correlated with packet volume
Variation of Number of Source IPs
19
Other Figures
20
Summary of Passive Observation

• TCP dominates (99 of the TCP packets are
  TCP/SYN)
• Near uniformity among destinations
• Hit pattern sweeping or random
• Variation over time
• Considerable amount of ICMP traffic
• Smaller set of sources scan all possible IPs
• Most of spoofed IPs are in class A
• The sources are expecting replies!

21
Outline

• The Authors and their Problem Statements
• Objective Terminology
• The study and Network Telescope
• Measurement Methodology
• Passive Measurement
• Active Measurement
• Comments.

22
The Big Picture

• Monitor network traffic to understand/track
  Internet attack activities
• Monitor incoming traffic to unused IP space

Internet
Unused IP space
Local network
23
Network Telescope

• Use a honeypot to keep conversation going
• (in the paper they used HoneyD and Active Sink)
• Answer PING
• Establish TCP connections
• Reply to application (e.g., HTTP) requests
• Till we find out what the intention is

24
Key Components
Responding to Application Requests
Filter
Taming the Traffic Volume
Analyzing Traffic Semantics
25
Measurement Methodology(Application-Level
Responders)

• Data-driven
• Which responders to build is based on observed
  traffic volumes
• Application-level Responders
• Not only adhere to the structure of the
  underlying protocol, but also to know what to say
• New types of activities emerge over time,
  responders also need to evolve

26
Radiation Activity Classification
Which Malware is Most Active?
What is the most Popular Application?
Which Vulnerability is Most Targeted?
27
A Rich Collection of Applications are targeted
in the Background Radiation

• Windows RPC
• HTTP
• Netbios/CIFS/SMB
• Virus backdoors (MyDoom, Beagle, etc.)
• Dameware
• Universal PnP
• Microsoft SQL (Slammer)
• MySQL
• DNS
• BitTorrent

28
TCP Port 80 (HTTP)

• Targeted against Microsoft IIS server.
• Dominant activity is a WebDAV buffer-overrun
  exploit.

29
TCP Port 80 (HTTP)
Port 80 Activities
30
Other Figures
31
Summary of Active Observation

• Study dominant activities on the popular ports
• Same Attacker on multiple networks
• Some sources avoid Class A
• Traffic is divided by ports
• Consider all connections between a
  source-destination pair on a given destination
  port
• Background Radiation concentrates on a small
  number of ports
• Only look at the most popular ports.
• Many popular ports are also used by the normal
  traffic
• ? use application semantic level.
• Many replies are needed to see what is happening

32
Conclusion

• Background Radiation is
• Complex in Structure, highly automated,
  frequently malicious, potentially adversarial
  matured in rapid speed
• Passive measurement reveal only part of the story
  
• Need to interact with the traffic to see what are
  the actual objectives of the attacker

33
Strengths

• First attempt to characterize background
  radiation
• Good Measurement Methodology
• Detailed set of active responders for popular
  ports.
• Meaningful Data Analysis
• Passive Analysis activities concentrate on
  popular ports.
• Active Analysis Extreme dynamism in many aspects
  of background radiation.

34
Weaknesses

• The filtering could be biased.
• The same kind of activity to all destination IP
  addresses.
• Fail to capture multi-vector worms that pick one
  exploit per IP address
• Significant amount of connections didnt proceed
• DHCP problem makes source IP address less
  accurate as source identity.
• To what extent the development of
  application-level responders can be automated?

35
Reference Back up Slide
36
References

• Barford2004 Paul Barford. Trends in Internet
  Measurement. PPT from U. of Wisconsin, Fall 2004
• MVS01 Moore, Geoffrey M. Voelker, and Stefan
  Savage. Inferring Internet Denial-of-Service
  Activity. In Proceedings of the 10th USENIX
  Security Symposium, pages 9--22. USENIX, August
  2001
• Google Earth

37
Measurement Methodology(Experimental Setup)

• Two different systems iSink, and LBL Sink.
• Traces collected from three sites
• Class A network
• UW campus
• Lawrence Berkeley Lab (LBL)
• Same forms of application response.
• Different underlying mechanisms.
• Support two kinds of data analysis
• Passive analysis no filter, no responder
• Active analysis with filter, and responder

38
Experimental Setup iSink
39
Experimental Setup LBL Sink
40

0 1 1 0 1 1 1 0 1 1
0 1
0 1 1 0 1 1
1 0 1 1 0 1
1 1 1
0 1 1
0101010 0101010 0101010 0101010 0101010 0101010
0101010 0101010 0101010 0101010 0101010 0101
0101010 0101010 0101010 0101010 0101010 0101010
0101010 0101010 0101010 0101010 0101010 0101010
0101010 0101010 0101010 0101010 0101010 0101010
0101010 0101010 0101010 0101010 0101010 0101010
0101010 0101010 0101010 0101010 0101010 0101010
0101010 0101010 0101010 0101010 0101010 0101010
0101010 0101010 0101010 0101010 0101010 0101010
0101010 0101010 0101010 0101010 0101010 0101010
0101010 0101010 0101010 0101010 0101010 0101010
0101010 0101010 0101010 0101010 0101010 0101010
0101010 010 0101010 0101010 0101010 0101010
0101010 0101010 0101010 0101010 0101010 0101010
0101010 0101010 0101010 0101010 0101010 0101010
0101010 0101010 0101010 0101010 0101010
01011111101111111010101111111001111111110101010
0101010 0101010 0101010 0101010 0101010 0101010
0101010
0 1 1 0 1 1 1 0 1 1
0 1
0 1 1 0 1 1
1 0 1 1 0 1
1 1 1
0 1 1
0 1 1 0 1 1 1 0 1 1
0 1
0 1 1 0 1 1 1 0 1 1
0 1
0 1 1 0 1 1 1 0 1 1
0 1
Thank You
0 1 1 0 1 1 1 0 1 1
0 1
0 1 1 0 1 1
1 0 1 1 0 1
1 1 1
0 1 1