OverTheWire Bandit: Level 8 → 9 Walkthrough

In the previous level we searched for a line by its content: the one with the word millionth. Now there’s no word to search for. The only clue is how many times the line appears: every other line is repeated, and the password is the only one that shows up just once. grep won’t help us here. Time to meet another classic pair: sort and uniq.

Passwords are shown as [REDACTED], following the OverTheWire rules.

Connecting to level 8

$ sshpass -p '[REDACTED]' ssh bandit8@bandit.labs.overthewire.org -p 2220
bandit8@bandit:~$ export TERM=xterm

The challenge

According to the level 9 page, the password is stored in the file data.txt and is the only line that occurs only once.

bandit8@bandit:~$ ls
data.txt

The solution: sort + uniq -u

bandit8@bandit:~$ cat data.txt | sort | uniq -u
[REDACTED]

Three commands chained with pipes, just like in the previous level. Let’s see what each one contributes:

Part	What it does
cat data.txt	Dumps the contents of the file.
sort	Sorts the lines alphabetically, so repeated lines end up next to each other.
uniq -u	Shows only the lines that have no copy right next to them (unique).

Why do we need sort?

This is the key detail of the level: uniq only compares each line with the one right before it. It doesn’t walk the whole file remembering what it has seen; it only looks at its neighbour. So if two copies of the same line are separated by other lines, uniq treats them as different.

A simple example, with a made-up five-line file:

pear
apple
pear
kiwi
apple

Here, uniq -u without sorting would print all five lines, because none of them has a copy right next to it. But if we sort first:

apple
apple
kiwi
pear
pear

Now the repeated lines are together, and uniq -u leaves only kiwi, the one that appears once. Exactly what we need with data.txt.

Rule of thumb: almost every time you use uniq, put a sort in front of it. Without sorting, the result depends on the order of the file and is rarely what you expect.

uniq options

Command	What it shows
uniq	One copy of each line (removes consecutive duplicates)
uniq -u	Only the lines that appear once
uniq -d	Only the lines that are repeated (one copy of each)
uniq -c	Each line prefixed with the number of times it appears

Other ways to do it

1. Without cat

bandit8@bandit:~$ sort data.txt | uniq -u
[REDACTED]

Just like with grep in the previous level, sort accepts the file as an argument, so cat isn’t needed. Same result, one process fewer.

2. Counting occurrences with uniq -c

bandit8@bandit:~$ sort data.txt | uniq -c | sort -n | head -n 1

This route is longer, but it shows you what’s inside the file:

• uniq -c prefixes each line with how many times it appears.
• sort -n sorts by that number numerically (without -n, sort would compare as text and 10 would come before 9).
• head -n 1 keeps the first line: the one with the lowest count, i.e. the one that appears only once, with a 1 in front of it.

If you drop the head, you’ll also see how many times each of the other lines is repeated, which is a nice way to understand how the challenge was built.

3. With awk alone, no sorting

bandit8@bandit:~$ awk '{c[$0]++} END {for (l in c) if (c[l] == 1) print l}' data.txt
[REDACTED]

Back to awk, which we met in the previous level, but this time using an associative array (a dictionary):

• {c[$0]++} runs on every line: it uses the whole line ($0) as the key and adds one to its counter.
• END {...} runs only once, after the whole file has been read.
• for (l in c) if (c[l] == 1) print l loops over every key and prints the ones whose counter is 1.

Unlike uniq, awk does remember every line it has seen, so it doesn’t need the file to be sorted.

Moving on to the next level

bandit8@bandit:~$ exit
$ sshpass -p '[REDACTED]' ssh bandit9@bandit.labs.overthewire.org -p 2220

Summary

Command / concept	What it’s for
sort	Sort lines, putting repeated ones together
sort -n	Sort numerically
uniq -u	Show only the lines that appear once
uniq -c	Count how many times each line appears
uniq -d	Show only the repeated lines
head -n 1	Keep only the first line
awk with arrays	Count occurrences without sorting
uniq without sort	Only compares consecutive lines: it almost always needs sort first