Investigating a computer breach is like investigating a crime scene. Detectives collect evidence, note any strange clues, and take inventory on loss and damage. An analysis of a computer compromise can either be done as the attack is happening or post-mortem.
Although it is unwise to trust any system log files on an exploited system, there are other forensic utilities to aid in the analysis. The purpose and features of these tools vary, but they commonly create bit-image copies of media, correlate events and processes, show low level file system information, and recover deleted files whenever possible.
It is also a good idea to record of all of the investigatory actions executed on a compromised system by using the script command, as in the following example:
script -q <file-name>
Replace <file-name> with file name for the script log. Always save the log file on media other than the hard drive of the compromised system — a floppy disk or CD-ROM works particularly well for this purpose.
By recording all your actions, an audit trail is created that may prove valuable if the attacker is ever caught.
Creating a bit-image copy of media is a feasible first step. If performing data forensic work, it is a requirement. It is recommended to make two copies: one for analysis and investigation, and a second to be stored along with the original for evidence in any legal proceedings.
You can use the dd command that is part of the coreutils package in Red Hat Enterprise Linux to create a monolithic image of an exploited system as evidence in an investigation or for comparison with trusted images. Suppose there is a single hard drive from a system you want to image. Attach that drive as a slave to the system and then use dd to create the image file, such as the following:
dd if=/dev/hdd bs=1k conv=noerror,sync of=/home/evidence/image1
This command creates a single file named image1 using a 1k block size for speed. The conv=noerror,sync options force dd to continue reading and dumping data even if bad sectors are encountered on the suspect drive. It is now possible to study the resulting image file or even attempt to recover deleted files.
The topic of digital forensics and analysis itself is quite broad, yet the tools are mostly architecture specific and cannot be applied generically. However, incident response, analysis, and recovery are important topics. With proper knowledge and experience, Red Hat Enterprise Linux can be an excellent platform for performing these types of analysis, as it includes several utilities for performing post-breach response and restoration.
Table 10-1 details some commands for file auditing and management. It also lists some examples that can be used to properly identify files and file attributes (such as permissions and access dates) to allow the collection of further evidence or items for analysis. These tools, when combined with intrusion detection systems, firewalls, hardened services, and other security measures, can help reduce the amount of potential damage when an attack occurs.
For detailed information about each tool, refer to their respective man pages.
|dd||Creates a bit-image copy (or disk dump) of files and partitions. Combined with a check of the md5sums of each image, administrators can compare a pre-breach image of a partition or file with a breached system to see if the sums match.||dd if=/bin/ls of=ls.dd |md5sum ls.dd >ls-sum.txt|
|grep||Finds useful string (text) information inside files and directories as well as reveals permissions, script changes, file attributes, and more. Used mostly as a piped command of for commands like ls, ps, or ifconfig.||ps auxw |grep /bin|
|strings||Prints the strings of printable characters within a file. It is most useful for auditing executables for anomalies such as mail commands to unknown addresses or logging to a non-standard log file.||strings /bin/ps |grep 'mail'|
|file||Determines the characteristics of files based on format, encoding, linked-libraries (if any), and file type (binary, text, and more). It is useful for determining whether an executable such as /bin/ls has been modified using static libraries, which is a sure sign that the executable has been replaced with one installed by a malicious user.||file /bin/ls|
|find||Searches directories for particular files. It is a useful tool for searching the directory structure by keyword, date and time of access, permissions, and more. It can also be useful for administrators that perform general system audits of particular directories or files.||find -atime +12 -name *log* -perm u+rw|
|stat||Displays file status information, including time last accessed, permissions, UID and GID bit settings, and more. It can be useful for checking when a breached system executable was last used or modified.||stat /bin/netstat|
|md5sum||Calculates the 128-bit checksum using the md5 hash algorithm. Use this command to create a text file that lists all crucial executables that are often modified or replaced in a security compromise. Redirect the sums to a file to create a simple database of checksums and then copy the file onto a read-only medium such as CD-ROM.||md5sum /usr/bin/gdm >>md5sum.txt|
Table 10-1. File Auditing Tools