Labs

Northstar

A backup exposure gives user access, and an outdated kernel combined with weak patch hygiene enables root escalation.

Northstar was the most realistic of the set for me. The foothold came from ==blue:weak file governance==, and root came from ==red:patch management debt==.

!![amber] Nothing flashy here; just layered operational failures compounding into root.

Reconnaissance

nmap -sC -sV -Pn -p- 10.10.11.89 -oN nmap-northstar.txt
Nmap scan report for 10.10.11.89
Host is up (0.13s latency).
Not shown: 65532 closed tcp ports (reset)

PORT   STATE SERVICE VERSION
22/tcp open  ssh     OpenSSH 7.9p1 Debian 10+deb10u4
80/tcp open  http    nginx 1.14.2
111/tcp open  rpcbind 2-4 (RPC #100000)

No obvious web vulnerability appeared immediately, so I switched to hunting ==amber:exposed artifacts==.

I also checked whether RPC service exposure translated into easy shares:

showmount -e 10.10.11.89
rpcinfo -p 10.10.11.89

rpcbind was reachable, but there were no useful exported paths, so web artifact discovery stayed primary.

Initial Access (Sanitized)

gobuster dir -u http://10.10.11.89 -w /usr/share/wordlists/dirb/common.txt -x txt,bak,tar.gz
/backup/                    (Status: 301)
/backup/site-2025-09.tar.gz (Status: 200)

The archive contained configuration data with credentials that were still valid for local access, which became the ==blue:initial access pivot==.

sam@northstar:~$ whoami
sam

Once inside, I captured quick baseline details before escalation checks:

id
hostnamectl
ls -lah /home/sam

Privilege Escalation (Sanitized but Insightful)

I checked common local vectors first:

sudo -l
find / -perm -4000 -type f 2>/dev/null
getcap -r / 2>/dev/null

No direct sudo or SUID path looked usable. Kernel triage looked more promising:

uname -a
Linux northstar 5.10.0-8-amd64 #1 SMP Debian 5.10.46-4 x86_64 GNU/Linux

linpeas highlighted this kernel generation as potentially affected by known local privilege escalation classes depending on patch level. I validated patch status manually and used a lab-safe public PoC workflow conceptually.

!![red] Kernel patch lag is a privilege escalation risk, not just a compliance issue.

I confirmed the branch by checking package patch levels and distro release metadata:

cat /etc/os-release
dpkg -l | grep linux-image | head

At that point, kernel path validity was stronger than any file-permission or sudo route on this host. I used a controlled local-kernel escalation workflow in the lab and verified root access.

This machine also reinforced a file-permission lesson: the initial backup archive should never have been web-readable. That single permission mistake created the entire path.

Proof (Sanitized)

whoami
root
cat /root/root.txt
flag output intentionally omitted to preserve the lab challenge

Key Takeaways

  • Backup exposure is a file-permission problem with direct security impact.
  • Kernel version and patch posture should be standard escalation checks.
  • When common vectors fail, disciplined enumeration still surfaces viable paths.
  • Defensive patching and artifact access control break this chain early.