CNIT 152: 6 Scoping & 7 Live Data Collection
2 likes602 views
The document outlines procedures for incident response, focusing on discovering the scope of incidents and gathering evidence. It details methods for preliminary data collection, the importance of using independent evidence sources, and strategies for analyzing incidents like data loss and fraud. It also emphasizes the use of live data collection while minimizing changes to systems and discusses best practices for forensic evidence gathering.
CNIT 152: 6 Scoping & 7 Live Data Collection
- 1. CNIT 152: Incident Response 6. Discovering the Scope of the Incident Updated 9-24-18
- 3. Examining Initial Data • Look at original alert • You may notice more than the person who reported it did • Ask about other detection systems and review what they recorded • Network administrators may not think like investigators • Gather context for the detection event
- 4. Preliminary Evidence • Determine what sources of preliminary evidence may help • Decide which sources you will actually use • Collect and review the evidence • Identify sources that are easy to analyze, and that quickly provide initial answers
- 6. Independent Sources • Firewall logs don't depend on registry keys, etc. • Multiple independent evidence sources lead to more reliable conclusions • It's difficult for an attacker to remove or modify evidence from all sources • Less likely that a routine process would overwrite or discard all evidence • Cross-check information, like date and time
- 7. Review • Attacker may be causing more damage • Test a detection method on one system, or a small date range of log entries • Make sure your detection method is fast and effective
- 8. Determining a Course of Action
- 9. Determining a Course of Action
- 11. Data Loss Scenario • Large online retailer • You work in IT security department • Customers are complaining about spam after becoming a new customer
- 12. Finding Something Concrete • Anecdotal customer complaints are inconclusive • Options • Work with customers and review their email • Reliability and privacy problems • Create fake customer accounts with unique email addresses
- 13. Fake Accounts • Use 64-character random usernames • Unlikely that spammers would guess the usernames • Monitor those accounts
- 14. Preliminary Evidence • Assuming customer data is being lost somehow: • Find where customer data is and how it is managed • One internal database on production server • One external database at a third-party marketing firm
- 17. Progress
- 18. Theories & Simple Tests • Insider • No easy way to test • Modified code on website to capture email addresses • Enter some fake accounts directly into database, bypassing the web form • Someone copying backup tapes • Add some fake accounts to the backups
- 19. Two Weeks Later • Spam comes to the first set of fake accounts • And to the accounts manually entered into the database • Suggests the website is not part of the problem • No spam from the accounts on the backup tape • Backups aren't the source of data loss
- 20. New Theories • Direct access to the database • Malware on the database server • Accessing it over the network
- 21. Monitoring Queries • Network-level packet captures • Expensive, powerful system required • If queries are encrypted, or malware is obfuscating them, it may be hard to decode the traffic • Database-level query monitoring and logging • Most efficient and reliable technique
- 22. Next Steps • Create a few more fake accounts • Talk to database and application administrators • To find out where data is stored • Scan through logs to see what is "normal" • No queries or stored procedures perform a bulk export of email addresses on a daily basis
- 23. Two Weeks Later • New accounts get spam • Retrieve query logs from time accounts created to spam time • For the field "custemail" • A single query is found • SELECT custemail FROM custprofile WHERE signupdate >= "2014-02-16"
- 24. Query Details • Feb 17, 2014 at 11:42 am GMT • Originated from IP in graphics arts dept. • Query used an database administrator's username from the IT dept. • Interview reveals that graphics arts dept. has no direct interaction with customers, only outside vendors
- 25. Leads
- 28. Results from Workstation • Examine images, focusing on actions at the time of the query • Malware found on workstation • Persistent, provides remote shell, remote graphical interface, ability to launch and terminate processes • Connects to a foreign IP • Has been installed for two years • Cannot determine how system was originally compromised
- 29. Final Steps of Investigation
- 30. Scoping Gone Wrong • After complaints from customers • Search every computer in the company for unique strings in the customer data, • And files large enough to include all the customer records
- 31. Problems • No evidence that the stolen data is stored on company servers • No evidence that the data is all being stolen at once in a large file • And even so, it would probably be compressed • Large amount of effort; low chance of success
- 32. Another Unwise Path • Focus on insiders • Who had access and knowledge to steal the customer data • Compile profiles of numerous employees • Review personnel files • Background checks, surveillance software capturing keystrokes and screen images • Video surveillance installed
- 33. Problems • Leads to a "witch hunt" • Invades privacy of employees • Large effort, small chance of success
- 34. Another Unwise Path • Because the data resides on the database server • Image and analyze RAM from the database server to hunt for malware • Because the hard drives are massive and too large to investigate easily
- 35. Problems • Once again, they have jumped to a conclusion • And ignored other possibilities • Large effort, low chance of success
- 36. Scenario 2: Automated Clearing House Fraud
- 37. Scenario 2: Automated Clearing House Fraud • Bank called the CEO--they blocked an ACH transfer of $183,642.73 • To an account that was never used before • Flagged by their fraud prevention system • Transfer from CFO's account, but he says he never authorized it
- 38. Facts from the Bank
- 39. Preliminary Evidence • Firewall • Two weeks of logs • Examine this first • CFO's laptop computer • Live response, RAM, hard drive • But maybe other computers are involved
- 40. Firewall Logs • Look near the time the unauthorized transfer occurred (4:37 pm) • See who logged in prior to that • Two computers logged in via HTTPS, making a number of connections between 4:10 pm and 4:48 pm • From two IPs -- one is CFO's, other not immediately recognized
- 41. Two Immediate Tasks • Gather complete forensic evidence from CFO's computer • Live response, RAM, and hard disk • Because evidence is being lost as time passes • Track down the other IP address and decide what action is appropriate
- 42. DHCP Logs • Search for the time in question • Get MAC address from DHCP logs • It's the MAC of the CFO's laptop!
- 44. Recap
- 45. Theories
- 46. Open Office Space • CFO's office is in clear view of other workers • It's unlikely that someone could go into it unobserved
- 47. CFO's Computer • Recently installed persistent executable • Send it to a third-party analysis site • It's a variant of the Zeus banking malware
- 48. Final Steps of Investigation
- 49. Scoping Gone Wrong • There are no recent antivirus or IDS alerts • So you believe the security issue must be at the bank • Tell the bank to find the attacker and put them in jail
- 50. Problems • No attempt to validate the bank's original data • Company assumes that existing network security measures would detect a problem, if there was one • Assumption that a third-party can help you • While you wait, data on company systems is lost
- 51. Another Unwise Path • CEO believes that security measures are in place to prevent malware, so the CFO must have initiated the transfer • The CEO wants you to investigate the CFO and avoid tipping him (or her) off
- 52. Problem • No security measures are perfect, not even two- factor authentication • Also, that sort of investigation is outside your expertise, and should be referred to an outside contractor
- 54. CNIT 152: Incident Response 7. Live Data Collection
- 55. Purpose of Live Collection • Preserve volatile evidence that will further the investigation • Also collect log files and file listings • Get answers quickly • Minimize changes to the system • Avoid disrupting business, causing crashes, or destroying evidence
- 56. When to Perform Live Response
- 57. Risks of Live Response
- 58. Altering the Evidence • All live response changes the system • Purists don't like it • But the alternative is to lose all volatile data and get only a disk image • You can minimize changes, but not eliminate them
- 59. Selecting a Live Response Tool • Homegrown Microsoft DOS batch script (or bash) • Perl-based script • There are specialized live-response products, free and commercial
- 60. Factors to Consider • Is the tool generally accepted in the forensic community? • Does the solution address the common operating systems in your environment? • Tools that use OS commands should contain known good copies of those commands, not trust the local commands on the suspect system
- 61. Factors to Consider • Does the solution collect data that is important to have, in your environment? • How long does a collection take? • Recommended: less than an hour per system • Is the system configurable? • Is the output easily reviewed and understood?
- 62. What to Collect • Current running state of the system • Network connections • Running Processes • What happened in the past • File listings, system logs • Usually a higher priority
- 63. The Deep End • Some organizations always collect entire RAM contents, or hard disk images • Don't collect data that you can't effectively use or understand • Collect data you can really use to quickly determine the impact of the incident
- 64. Data to Collect
- 65. Data to Collect
- 66. Data to Collect
- 67. Complete RAM Capture • Requires specialized tools to collect and interpret • Not part of Live Response • Sometimes needed on carefully chosen systems
- 68. Collection Best Practices • Practice on a test system first • Learn how fast the process is, and how large the output is • Practice handling problems • Broken USB port or NIC • Locked screen
- 69. Caution: Malware • The system you are examining may be infected with malware • Any media you connect may become infected • Any credentials you use may be compromised
- 73. Horror Stories: IR Procedures • Copy live response toolkit to affected systems, save collected data back to that same system including full RAM dump, several gigabytes in size • Remotely log in with domain administrator account, run netstat & Task Manager • Pull out the plug, image the hard drive
- 74. Good Methods of Live Response • Network share on a dedicated file server • Not part of any domain • Used throwaway credentials not used for any other purpose • Two folders • Read-only containing the live response toolkit • Writeable for output from live response toolkit
- 76. Live Response Tips • Air-gap for evidence server • Logging and auditing access to evidence server • Automate process for consistency • Live Response software must run as Local Administrator/root
- 77. Media • Some computers cannot connect external media • Hardware failure, configuration, etc. • Common options for running toolkit • CD-ROM, DVD, network share • Encrypted network streaming tool like cryptcat or stunnel to send output to another system
- 78. Unexpected OS • Cannot run your normal live response toolkit • If you can't update or modify your toolkit to run • Perform manual live response
- 79. Unexpected OS • Create a checklist of the automated steps in the toolkit for a similar OS • Research command-line options • Test them on a known clean system, if possible • Manually perform steps to collect evidence
- 80. Automation • Decreases human error • Makes processes more consistent and faster • Helps to prevent bad guys from gathering intelligence about how you respond to incidents • Anything you do on the evidence system may be sent to the bad guys
- 82. Live Data Collection on Microsoft Windows Systems
- 83. Three Main Options • Use a prebuilt kit (Mandiant Redline) • Create your own • Use a hybrid of the two
- 84. Mandiant Redline • Install the Redline MSI package on a trusted workstation • Create the Redline collector on the trusted system • The only step you take on a suspect system is to run the stand-alone Redline collector batch script • Automatically saves data to the same location you ran the script from
- 85. Do It Yourself • Make your own live response toolkit • Decide what OS to support • Windows has many versions, and big differences between 32-bit and 64-bit • Find tools that collect the information you want
- 86. Windows Built-in Tools • Copy these files from a clean Windows system • Also copy cmd.exe • "Trusted binaries" • Don't trust files on the evidence machine
- 87. Free Tools • Use command-line versions, not GUI versions • Easier to script • Less impact • Rename every tool so you can identify it as something you added to the system • Prepend "t_"
- 88. Other Data Items • Prefetch information • System restore point information • Browser history, and more • Balance your needs with the impact the collection has on the system
- 89. Scripting Language • Choose one • MS-DOS Batch (lowest impact) • VBScript • Perl • Python
- 90. Scripting Tips • Add logging and compute a checksum of collected data • Be careful with file and directory names • They may be long or include spaces • Test your script extensively • Built a test environment that resembles your production systems • Watch for errors and unexpected results
- 91. Memory Collection • Tools for a full memory dump • AccessData FTK Imager Lite • Mandiant Memoryze • Monsools Windows Memory Toolkit
- 92. Mandiant Memoryze • Command-line tool: MemoryDD.bat
- 93. AccessData FTK Imager Lite
- 94. Individual Process RAM Dump • Tools • Mandiant Memoryze • Microsoft userdump • Microsoft procdump • Ntsecurity.nu pmdump
- 96. Live Data Collection on Unix-Based Systems
- 97. LINReS • From Network Intelligence India • Written for RedHat 3 and 4, not updated since 2006 • Useful mainly as an example to guide you in making a custom tool
- 98. Do It Yourself • Really the only option • Make some scripts • Mandiant uses Bourne shell scripts that make scripts for various Unix/Linux versions • Requires constant maintenance
- 99. Language Choices • Perl • Python • Bourne shell • BASH (Mandiant uses this) • others
- 100. Apple Systems • system_profiler • Very long list of software, hardware, logs, etc.
- 101. system_profiler
- 102. system_profiler
- 103. Built-in Unix Tools
- 104. Built-in Unix Tools
- 105. Built-in Unix Tools
- 106. Memory Collection • The memory device is handled differently in every version of Unix, BSD, and Linux • In earlier versions, you could just use dd to collect RAM through the /dev/mem device • Direct access to memory os now blocked for security reasons • Use LiME – Linux Memory Extractor (link Ch 7c)
- 107. Loadable Kernel Modules • LiME is an LKM • Must be compiled for the exact kernel version running on the target system • No ability to include checksums of the output • You must do that yourself
- 108. Collection from BSD-Based Kernels • Use dc3dd or dcfldd to capture contents of /dev/mem • They are like dd but also include checksums • In recent versions, there's no End Of File mark in /dev/mem, so you must manually specify how many bytes to capture
- 109. Collection from Apple OS X • Memoryze for Mac (link Ch 7d) • Mac Memory Reader seems to be gone
- 110. Individual Process Dump • "gcore" (part of gdb, the GNU debugger)