Signal9Signal9 — home
online
back to labs

Metadata Inventory: The File That Lies

MediumDigital Forensics & Incident Response15-30 min
The brief

You're triaging the contents of /evidence/disk recovered from a workstation. Build a defensible inventory — names, sizes, and hashes — but the real goal is the anomaly hunt: one of these files is not what its name claims. An attacker likes to hide an executable behind an innocent extension so it slips past a glance at the directory listing. The directory entry and the extension are just labels; the file's magic bytes are the truth — and a matching extension is no proof either, so check every file. Find the disguised file.

Produce two files: /root/case/inventory.txt (every evidence file's name with its SHA-256 — sha256sum's own <digest> <path> lines are fine; grading checks each file's hash against the pristine evidence) and /root/case/suspect.txt (just the disguised file's name, one line — the bare name, its full path, or file's own line for it are accepted; a list of several files is not).

What you'll do
  • Build a metadata inventory of /evidence/disk (name + size + SHA-256) and save it to /root/case/inventory.txt
  • Use `file` to compare each file's real type to its extension
  • Record the name of the file whose true type does not match its extension to /root/case/suspect.txt

Hands-on lab: you work on a real Linux box in your browser, graded on the machine's actual state.