cheatsheet

bzip2 Command in Linux: Syntax, Options & Practical Examples

bzip2 compresses single files with the Burrows–Wheeler algorithm. By default it replaces the original with a .bz2 file; use -k to keep the source, or pipe to tar for directory archives.

Published

Updated

Read time 7 min read

Reviewed byDeepak Prasad

bzip2 Command in Linux: Syntax, Options & Practical Examples
About bzip2 compresses single files with the Burrows–Wheeler algorithm. By default it replaces the original with a .bz2 file; use -k to keep the source, or pipe to tar for directory archives.
Tested on Ubuntu 25.04 (Plucky Puffin); bzip2 1.0.8; kernel 7.0.0-27-generic
Package bzip2
Man page bzip2(1)
Privilege user
Distros

Most Linux distros ship bzip2 in the base or main repository.

Faster compression: gzip. Directories: archive with tar first (tar cjf archive.tar.bz2 dir/).

bzip2 — quick reference

Compression

Compress one or more files. Default action removes the original unless you pass -k.

When to use Command
Compress a file (replaces original with .bz2) bzip2 file.txt
Compress and keep the original file bzip2 -k file.txt
Overwrite an existing .bz2 without prompting bzip2 -f file.txt
Compress several files in one invocation bzip2 file1 file2
Write compressed data to stdout (keep original on disk) bzip2 -c file.txt > file.txt.bz2
Force compression even if invoked as bunzip2 bzip2 -z file.txt
Fastest block size (less compression) bzip2 -1 file.txt
Best compression (slowest, largest blocks) bzip2 -9 file.txt
Use less RAM on memory-constrained hosts bzip2 -s file.txt

Decompression

When to use Command
Decompress and remove the .bz2 file bzip2 -d file.txt.bz2
Same as -d when called as bunzip2 bunzip2 file.txt.bz2
Decompress to stdout (original .bz2 stays) bzip2 -dc file.txt.bz2
Same stdout decompress as bzcat bzcat file.txt.bz2

Integrity and messaging

When to use Command
Test archive integrity without decompressing bzip2 -t file.txt.bz2
Print compression ratio and file names bzip2 -v file.txt
Suppress non-critical messages bzip2 -q file.txt

Help and version

When to use Command
Show usage summary bzip2 -h
Show version and license bzip2 -V

bzip2 — command syntax

bzip2 reads files or stdin and writes .bz2 output. Synopsis from bzip2 --help on Ubuntu 25.04 (bzip2 1.0.8):

text 
bzip2 [flags and input files in any order]

  -d --decompress     force decompression
  -z --compress       force compression
  -k --keep           keep (don't delete) input files
  -f --force          overwrite existing output files
  -t --test           test compressed file integrity
  -c --stdout         output to standard out
  -q --quiet          suppress noncritical error messages
  -v --verbose        be verbose (a 2nd -v gives more)
  -s --small          use less memory (at most 2500k)
  -1 .. -9            set block size to 100k .. 900k

bzip2 compresses files only — not directory trees. Use tar to archive folders (tar cjf backup.tar.bz2 /path).

bzip2 — command examples

Essential Compress a file (original removed)

Default compression replaces file.txt with file.txt.bz2.

Run the command:

bash 
WORKDIR=/tmp/bzlab
mkdir -p "$WORKDIR" && cd "$WORKDIR"
echo "compress me please" > file.txt
bzip2 file.txt
ls -la

Sample output:

text 
total 4
drwxr-xr-x  2 root root  60 Jul  1 18:06 .
drwxrwxrwt 17 root root 400 Jul  1 18:06 ..
-rw-r--r--  1 root root  56 Jul  1 18:06 file.txt.bz2

Only the .bz2 remains unless you used -k.

Essential Keep the source file with -k

Use -k when scripts or teammates still need the uncompressed copy.

Run the command:

bash 
WORKDIR=/tmp/bzlab
mkdir -p "$WORKDIR" && cd "$WORKDIR"
echo "compress me please" > file.txt
bzip2 -k file.txt
ls -la file.txt*

Sample output:

text 
-rw-r--r-- 1 root root 19 Jul  1 18:06 file.txt
-rw-r--r-- 1 root root 56 Jul  1 18:06 file.txt.bz2

Both files coexist after compression.

Essential Verify a .bz2 without extracting

Run -t after transfers or before deleting the only copy of an archive.

Run the command:

bash 
WORKDIR=/tmp/bzlab
mkdir -p "$WORKDIR" && cd "$WORKDIR"
echo "compress me please" > file.txt
bzip2 -k file.txt
bzip2 -t file.txt.bz2
echo "exit code: $?"

Sample output:

text 
exit code: 0

A non-zero exit code means the archive is corrupt or truncated.

Common Read compressed data on stdout with -dc

Pipe decompressed bytes into another tool without writing an intermediate file.

Run the command:

bash 
WORKDIR=/tmp/bzlab
mkdir -p "$WORKDIR" && cd "$WORKDIR"
echo "compress me please" > file.txt
bzip2 -k file.txt
bzip2 -dc file.txt.bz2

Sample output:

text 
compress me please

bunzip2 -c and bzcat behave the same on this build.

Common Compress several files at once

Each input file becomes its own .bz2 — bzip2 does not bundle multiple files into one stream.

Run the command:

bash 
WORKDIR=/tmp/bzlab
mkdir -p "$WORKDIR" && cd "$WORKDIR"
echo "data" > f1.txt
echo "more" > f2.txt
bzip2 -k f1.txt f2.txt
ls *.bz2

Sample output:

text 
f1.txt.bz2
f2.txt.bz2

For one archive of a directory, use tar cjf instead.

Common Trade speed for ratio with -1 and -9

Block sizes range from 100k (-1) to 900k (-9). Higher levels compress more but take longer.

Run the command:

bash 
WORKDIR=/tmp/bzlab
mkdir -p "$WORKDIR" && cd "$WORKDIR"
echo "compress me please" > orig.txt
bzip2 -1 -k -f orig.txt
bzip2 -9 -k -f orig.txt
ls -la orig.txt.bz2

Sample output:

text 
-rw-r--r-- 1 root root 56 Jul  1 18:06 orig.txt.bz2

On tiny files the ratio difference is negligible; -9 shines on larger text logs.

Advanced See compression statistics with -v

Verbose mode prints space saved — helpful when comparing levels on real data.

Run the command:

bash 
WORKDIR=/tmp/bzlab
mkdir -p "$WORKDIR" && cd "$WORKDIR"
echo "compress me please" > orig.txt
bzip2 -v -k -f orig.txt

Sample output:

text 
orig.txt:  0.339:1, 23.579 bits/byte, -194.74% saved, 19 in, 56 out.

Negative "saved" on very small files is normal — the .bz2 header can be larger than the payload.

Advanced Decompress with bunzip2

bunzip2 is the same binary with decompression as the default action.

Run the command:

bash 
WORKDIR=/tmp/bzlab
mkdir -p "$WORKDIR" && cd "$WORKDIR"
echo "compress me please" > file.txt
bzip2 -k file.txt
bunzip2 -c file.txt.bz2

Sample output:

text 
compress me please

Use bzip2 -d in scripts when you want one command name for both directions.

bzip2 — when to use / when not

Use bzip2 when Use something else when
You need smaller .bz2 files than gzip on text-heavy data Speed matters more than ratio — try gzip
You are building tar.bz2 archives on Debian/Ubuntu workflows You need maximum compression and can wait — consider xz
You are compressing individual log or CSV files You must compress a directory tree — use tar first
You want a well-supported format on older systems You need random access inside the archive — zip or uncompressed tar

bzip2 vs gzip

bzip2 gzip
Typical ratio on text Better (smaller .bz2) Faster, slightly larger .gz
Speed Slower compress/decompress Faster
Default level Block size 6 (900k class) Level 6
Directories No — files only No — files only

Both pair with tar for tar.bz2 and tar.gz archives.

Compression and archiving tools often used beside bzip2.

Command One line
bzip2 Block-sort file compressor (this page)
tar Archive files and directories before compression
lbzip2 Parallel bzip2 implementation

Browse the full index in our Linux commands reference.

bzip2 — interview corner

What is bzip2 used for in Linux?

bzip2 compresses one file at a time using the Burrows–Wheeler transform and Huffman coding. The default action replaces file with file.bz2 and deletes the original unless you pass -k.

It does not archive directories — admins combine it with tar (tar cjf backup.tar.bz2 /data) for folder backups.

A strong answer is:

"bzip2 compresses individual files to .bz2 with good text ratios; I use -k to keep the original and tar for directory trees."

How does bzip2 compare to gzip?

bzip2 usually achieves better compression on text at the cost of CPU time. gzip compresses and decompresses faster with slightly larger output.

Choose bzip2 when disk or bandwidth is tight; choose gzip when throughput or low latency matters.

A strong answer is:

"bzip2 trades speed for smaller files; gzip is faster. For directories I tar first, then pick bzip2 or gzip on the archive."

How do you check whether a .bz2 file is valid?

Run bzip2 -t archive.bz2. Exit code 0 means the blocks decompress cleanly; non-zero indicates corruption. The test does not write an uncompressed file to disk.

A strong answer is:

"bzip2 -t verifies integrity without extracting — I run it after downloads or before deleting the only uncompressed copy."

How do you decompress to stdout without removing the .bz2 file?

Use bzip2 -dc file.bz2 or bzcat file.bz2. The -c / stdout mode writes decompressed bytes to the terminal or the next pipe stage while leaving file.bz2 on disk.

A strong answer is:

"bzip2 -dc or bzcat streams decompressed data to stdout — useful in pipelines like bzip2 -dc log.bz2 | grep ERROR."

What do bzip2 -1 through -9 control?

They set the compression block size from 100k (-1, --fast) up to 900k (-9, --best). Higher levels spend more CPU and memory for potentially smaller output.

A strong answer is:

"bzip2 -1 is fastest with smaller blocks; -9 uses 900k blocks for best ratio. I default to the standard level unless profiling shows a need to change."

Troubleshooting

Symptom Likely cause Fix
bzip2: Input file is a directory bzip2 only handles files tar cjf archive.tar.bz2 directory/
Output file already exists .bz2 present without -f bzip2 -f or remove the old archive
bzip2: Data integrity error Truncated or corrupt file Re-download; verify with bzip2 -t
Compressed file larger than original Payload too small for header overhead Normal on tiny files; skip compression or use gzip
Can't guess original name Wrong extension or stdin without -c Pass the full .bz2 path or use bzip2 -dc
Rohan Timalsina

is a technical writer and Linux enthusiast who writes practical guides on Linux commands and system administration. He focuses on simplifying complex topics through clear explanations.