Modern Computer Viruses
Author - Chas Tomlin
- IT Security Specialist in School of Electronics and Computer Science
- Member of Shadowserver
- Maintain a db of currently ~15,000 unique viruses and their associated network traffic.
Aim
- Take a look at current viruses/malware and how they propagate
- See what features modern viruses come with
- See what impact an infected machine might have on your network
- Learn to detect and prevent viruses
Four Installments
- Infection
- Payload
- Impact
- Response
Focus on Windows viruses
Response
- Prevention
- Analysis
- Detection
- Reporting
- Prosecute
- Phishing
Prevention
- Anti Virus software
- Patching
- Least privilege

Sandboxing - Analysis
- Run a virus on a disposable system to observe its network behaviour.
- Can be a virtual environment such as VMware, qemu, xen etc.
- Malware may detect virtualisation
- Highly monitored
Method One
- API-hooking based technique to monitor the binary during runtime. Rootkit.
- Norman sandbox
- CWSandbox
Method Two
- Logging network traffic, file system changes, registry changes.
API Hooking output
<WINSOCK_SECTION>
<CONNECTIONS_UNKNOWN>
<CONNECTION ProbableWebProtocolType="Unknown" ConnectionEstablished="0" Socket="0">
<ACTION>Winsock initialized (WSAStartup)</ACTION>
<ACTION>GetHostByName for host "real80.act10l.com.ar" => 72.20.13.88</ACTION>
</CONNECTION>
</CONNECTIONS_UNKNOWN>
<CONNECTIONS_UDP />
<CONNECTIONS_LISTENING />
<CONNECTIONS_INCOMING />
<CONNECTIONS_OUTGOING>
<CONNECTION TransportProtocol="TCP" RemoteAddr="72.20.13.88" RemotePort="80"
ProbableWebProtocolType="IRC" ConnectionEstablished="1" Socket="700">
<USERNAME>[1]DEU|3479169</USERNAME>
<PASSWORD><real80.connectpass></PASSWORD>
<NICK>[1]DEU|3479169</NICK>
<ACTION>Joined channel ##http-scan##,##http-down1##,##http-down2## argrulex3</ACTION>
<ACTION>Joined channel ##http-scan##,##http-down1##,##http-down2## argrulex3</ACTION>
<ACTION>Joined channel ##http-scan##,##http-down1##,##http-down2## argrulex3</ACTION>
</CONNECTION>
</CONNECTIONS_OUTGOING>
<CONNECTIONS_OUTGOING_BLOCKED />
</WINSOCK_SECTION>
Detection
- Intrusion Detection
- Extrusion Detection
- Intrusion Prevention
- Darknet monitoring
- Sensible firewalling
IDS - Detection

Extrusion Detection
- Deals with detecting Intrusions by monitoring outbound traffic.
- Useful when firewalls have been circumvented (internal infection).
- Easily implemented on standard IDS.
- Detect malicious outbound IRC traffic using snort;
This rule merely looks for IRC traffic on any TCP port (by detecting NICK change events, which occur at the beginning of the session) and sets the is_proto_irc flowbit. It does not actually generate any alerts itself.
alert tcp any any -> any any (msg:"IRC TRAFFIC DETECTED BY NICK CHANGE"; flow: to_server,established; content:"NICK "; nocase; offset: 0; depth: 5; flowbits: set,is_proto_irc; flowbits: noalert; sid:9000075; rev:1;)
Darknet - Detection
- A Darknet is a portion of unused IP space
- No services exist in a Darknet
- Packets that enter are logged
- Can be IPv4 or IPv6
- Darknets can be used to detect;
- Scan traffic
- Infected hosts, worm traffic
Reporting
- Found a malicious web site? Report it!!!
- Use whois to find abuse contact
- Some ISPs are very responsive
Prosecute
- What are law enforcement doing? Not enough.
- Not enough resources/skills
- Not enough support from prosecution
- Some are prosecuted though

Phishing
- Fake web sites made to look like the real thing to steal your passwords
- Usually hosted on hacked servers
- Spam sent to lure the victim to the site.

- Very large proportion of phish sites are hosted on hacked Unix/Linux servers


- Experts believe that 50-75% of phishing attacks are due to one group of phishers. Rock phishers
- Banks are currently not tackling the problem as the cost of solutions far out weighs the cost of the problem.
- Two-factor authentication

- Even two-factor authentication can be susceptible to phish/man-in-the-middle attacks
<< | Viruses | >>


