Explore our new Cybersecurity Terms Glossary, explained by our experts.

View the Glossary View the Glossary
PRIVIA

Cybersecurity Terms Glossary

What Is Linux Forensics?

Linux forensics examines servers and workstations where the evidence is scattered across text logs, systemd units, shell history, and the proc filesystem rather than a central registry.

Illustration for the cybersecurity term Linux Forensics
Home / Blog / Cybersecurity Terms / What Is Linux Forensics?
17 November 2026

What Is Linux Forensics?

What Is Linux Forensics?

Most of the internet runs on Linux, so most web application compromises end with a Linux investigation. The experience is different from Windows work in one structural way: there is no registry. Configuration and history are spread across text files, binary journals, scheduler directories, and the kernel’s view of running processes, and knowing where to look is most of the skill.

Capture the Living System First

A compromised web server is usually still running when you arrive, and shutting it down discards the most valuable evidence. Capture memory with AVML or LiME before anything else, because an attacker’s payload may exist nowhere on disk. Then record volatile state: running processes with their full command lines, open network connections, loaded kernel modules, and the contents of /proc for anything suspicious.

That last point deserves emphasis. A process whose executable has been deleted from disk still has its binary available through /proc/<pid>/exe, and its working directory, open file descriptors, and memory maps are all readable. Recovering a deleted implant from a running process is routine and becomes impossible the moment someone reboots to “clear it up”.

Where the Logs Live

Authentication activity appears in /var/log/auth.log on Debian-derived systems and /var/log/secure on Red Hat derivatives: SSH logins, key-based authentications, sudo invocations, and account changes. The systemd journal holds the same material in binary form with more structure, and can be exported for timeline work. Login records in wtmp, failed attempts in btmp, and current sessions in utmp are binary files read with last, lastb, and utmpdump — and their truncation to zero bytes is a classic sign of clean-up, covered in finding traces of log deletion.

Shell history is useful and frequently misread. By default .bash_history carries no timestamps unless the environment sets a history format, it is written on clean exit rather than per command, and an attacker who uses a non-interactive shell leaves nothing in it at all. Treat its contents as evidence and its emptiness as meaningless.

Persistence Has Many Homes

Package Verification Finds Tampering Fast

Distributions record a hash for every file installed from a package, so rpm -Va or debsums will list files that no longer match. On a server built from packages and configuration management, that single command often identifies a trojanised binary in seconds. Combine it with inode and timestamp analysis: ext4 records a creation time that stat does not display but debugfs will, which sometimes contradicts a modification time an attacker has altered.

Collection and Tooling

UAC and CyLR gather triage artefacts across many hosts quickly, Velociraptor does the same at scale with live hunting, and plaso builds the super timeline. The Sleuth Kit handles the file system layer, and Volatility analyses the memory image, provided you have a profile matching the kernel.

Containers change the picture — see Docker container forensics — and so does cloud hosting, where a snapshot and an audit trail may be more productive than the guest operating system. Our investigation and incident response services cover Linux estates including the cases where the only surviving evidence is in memory on a host nobody wants to take offline.

FAQ – Frequently Asked Questions

Should I Shut Down a Compromised Linux Server?

Not before capturing memory and volatile state. Many Linux implants run entirely from memory, and a process whose file has been deleted from disk can still be recovered through the proc filesystem while it is running. Isolate the host at the network layer, capture RAM, record processes and connections, then decide about imaging or shutdown.

Is an Empty Bash History Suspicious?

It is a data point, not proof. History is written when a shell exits cleanly, so a terminated session leaves nothing, non-interactive and scripted commands are never recorded, and many service accounts legitimately have no history at all. A file truncated to zero bytes with a recent modification time is more interesting than a file that was never written.

What Is the Fastest Way to Find a Trojanised Binary?

Package verification. Both rpm and dpkg-based distributions record hashes for files installed from packages, so rpm -Va or debsums will flag anything that has changed. On a server built from packages and managed by configuration tooling, the list of mismatches is usually short enough to review by hand and frequently contains the answer.

How Do I Build a Timeline on Linux?

Combine file system metadata with log sources using plaso, which parses ext4 and XFS metadata, systemd journals, authentication logs, and web server logs into one chronology. Add the ext4 creation time read through debugfs where it matters, and record the system time zone, since mixed UTC and local time entries are a frequent source of error.

Related Terms

You May Be Interested In These

What Is Adversarial Machine Learning?

Cybersecurity Terms

17 November 2026

What Is Adversarial Machine Learning?

Read More
What Are Business Logic Flaws?

Cybersecurity Terms

17 November 2026

What Are Business Logic Flaws?

Read More
What Is Hybrid Key Exchange?

Cybersecurity Terms

17 November 2026

What Is Hybrid Key Exchange?

Read More