find Command in Linux: Search Files and Folders

Quick take: The find command searches a directory tree for files matching criteria such as name, type, size, or age, and can run an action on each match with -exec. The basic form is find PATH -name 'pattern'.

Introduction

The find command is the most powerful file-search tool in Linux. It walks a directory tree and tests each file against conditions you supply — name, type, size, modification time, ownership, permissions — and can then act on the matches, from deleting them to running any command.

This guide builds from simple name searches up to combined conditions and the powerful -exec action that turns find into a batch-processing engine.

Syntax

The basic syntax of the find command is:

find [PATH...] [EXPRESSION]

Common Options and Parameters

The most useful options and parameters for the find command:

Practical Examples

Real find commands you can run today:

# Find files by name under the current directory
find . -name '*.log'
# Case-insensitive name search
find /etc -iname 'nginx*'
# Find only directories
find /var -type d -name cache
# Files larger than 100 MB
find / -type f -size +100M
# Files modified in the last 24 hours
find . -mtime -1
# Delete .tmp files older than 7 days
find /tmp -name '*.tmp' -mtime +7 -delete
# Run a command on each match
find . -name '*.sh' -exec chmod +x {} \;
# Limit search depth to one level
find . -maxdepth 1 -type f

Advanced find: Combining Conditions

find evaluates conditions left to right and joins them with an implicit AND, but you can build precise queries with OR (-o), negation (!), and grouping with escaped parentheses.

# Files ending in .jpg OR .png
find . \( -name '*.jpg' -o -name '*.png' \)

# Files NOT owned by the current user
find /var/www ! -user $USER

# Large, old log files: over 50 MB AND older than 30 days
find /var/log -type f -size +50M -mtime +30

For actions, prefer -exec cmd {} + over \; when the command accepts many arguments — it batches files into far fewer invocations and runs dramatically faster on large trees.

Running find Safely

find can delete or modify thousands of files at once, so a careless command is dangerous. The golden rule is to preview before you act: run the find with no action and confirm the list, then re-run it with -delete or -exec. For deletions, find … -print first, eyeball the output, then swap in -delete.

Also be deliberate about where you start. find / … walks the entire filesystem and can be slow and noisy; scope the search to the relevant directory and add -maxdepth to limit recursion when you only need the top levels.

Tips and Best Practices

• Always quote wildcard patterns (-name '*.log') so the shell passes them to find instead of expanding them first.
• Test destructive searches without the action first — run the find alone, confirm the list, then add -delete or -exec.
• For very large trees, find ... -exec cmd {} + (note the +) batches arguments and runs far fewer processes than \;.

Final Thoughts

find turns “where is that file?” into a precise, scriptable query. Master the core tests — -name, -type, -size, -mtime — and the -exec action, and you can locate and process files across an entire system. It is one of the highest-leverage commands a Linux administrator can know.