Goal-only tasks, no walkthroughs — same format as the RHCSA practice exercises. You’re told the objective, not the commands; a real engagement doesn’t hand you the payload either. Run every one of these against the lab built in home-lab-setup — never against anything you don’t own or don’t have explicit written authorization to test, per rules-of-engagement-and-legal.
How to use this
- Snapshot the relevant target VM before each exercise. Several of these are meant to leave the target in a broken/compromised state — that’s the point, and you revert after, not before.
- Try to complete each task using the concepts in exploitation-concepts and the tools in common-tools before looking up a specific walkthrough for that exact CVE — the goal is transferable skill, not memorizing one target’s specific solution.
- Write a one-paragraph note after each exercise: what you found, how, and what the fix would be. This is the actual habit that separates “ran a tool” from “did a pentest” — see post-exploitation-and-reporting.
- Time yourself loosely once you’ve done a category once cold. Not to simulate exam pressure like the RHCSA drills — this is about noticing which categories are still slow for you, which tells you where to spend more repetition.
Reconnaissance & Scanning
-
Against Metasploitable2, determine every open TCP port and identify the service + version running on each, without yet trying to exploit anything.
Verify: you have a written list of port → service → version for at least 10 open ports. -
From that service list, identify which specific service(s) are running a version with a publicly documented, named vulnerability (not just “old” — an actual CVE or well-known exploit name).
Verify: you can name the vulnerability and cite where you confirmed it (searchsploit, a CVE database, exploit-db). -
Enumerate every share exposed by Metasploitable2’s file-sharing service and determine which, if any, are accessible without authentication.
Verify: a list of share names with their access level (readable/writable/none), gathered without guessing — actually enumerated.
Exploitation — Network Services
-
Gain a remote shell on Metasploitable2 using one of the vulnerabilities identified in exercise 2, via a purpose-built exploitation framework rather than hand-writing an exploit.
Verify: a working shell withwhoamireturning a result, and you can state which specific CVE/vulnerability you used. -
Once you have a shell, determine what privilege level you landed at, and — if it isn’t already root — escalate.
Verify:idshowsuid=0(root). -
Gain a shell using an entirely different vulnerable service than the one from exercise 4 (Metasploitable2 has several) — don’t reuse the same technique twice.
Verify: a second, independent shell via a different named vulnerability.
Password & Credential Testing
-
Metasploitable2 has at least one service accepting weak/default credentials over the network directly (not a hash to crack — a live login attempt). Find it and demonstrate a successful login using a wordlist-based attack, not a guessed password.
Verify: a successful authenticated session, plus the specific username/password pair the attack found. -
Extract a password hash from a compromised target (from exercise 4 or 6) and crack it offline.
Verify: a plaintext password recovered, plus which hash format you identified it as before choosing a cracking mode.
Web Application Testing (DVWA)
-
At DVWA’s Low security setting, find and demonstrate a SQL injection that extracts data the application was never meant to return (e.g., every username and password hash in the database), first by hand in the browser/Burp before reaching for an automated tool.
Verify: actual extracted data, plus the specific injected input that caused it. -
Confirm the same injection using an automated tool, and use it to enumerate the full database schema (table names, column names) — not just the one table from exercise 9.
Verify: a full list of tables and columns the tool extracted. -
Find a reflected or stored XSS vulnerability in DVWA and demonstrate impact beyond a plain
alert()popup (e.g., something that would exfiltrate a cookie/session token in a real scenario — you don’t need a live receiving server, describing the mechanism precisely counts).
Verify: a working payload plus a one-sentence explanation of what it would actually let an attacker do. -
Find a broken access control issue in DVWA — access another simulated user’s data or action by manipulating a request, without ever being given that access directly.
Verify: you can show the exact parameter/ID you changed and the unauthorized data/action it exposed. -
Repeat exercises 9 and 11 at DVWA’s Medium setting. Note specifically what changed in the application’s defenses and how you adapted the technique — not just that it eventually worked.
Verify: a short written comparison of the Low vs. Medium bypass.
Wireless (requires the external adapter from lab-setup-and-os-choice)
-
Against a WiFi network you own and control, capture a WPA2 4-way handshake and crack the pre-shared key offline using a wordlist you know contains it (i.e., prove the pipeline works before testing against an unknown/strong password).
Verify: the recovered key matches the network’s actual configured password. -
Determine whether that same network has WPS enabled, and if so, what that means for its practical security regardless of WPA2 password strength.
Verify: a clear yes/no on WPS status plus a one-sentence explanation of the risk if enabled.
Reporting
- Take exercises 4–13 (whichever you completed) and write them up as a short-form finding report: what was found, how it was found, actual demonstrated impact, and a concrete remediation for each — structured the way post-exploitation-and-reporting describes, not just a command transcript.
Verify: someone who wasn’t in the room could read the report and understand exactly what’s broken and how to fix it, without needing you to explain it verbally.
Hosted platforms — once the lab exercises feel routine
common-tools already points at HackTheBox and TryHackMe for legally-sanctioned targets beyond your own lab. The natural next step after this file: pick one beginner-track machine/room per category above (a “starting point” HTB box, a TryHackMe web-focused room) and repeat the same exercises against a target you didn’t build yourself — that’s the real test of whether the skill transferred or whether you’d just memorized Metasploitable2 and DVWA specifically.
Software engineering
Infrastructure & security
Computer science
Hardware & engineering