Back to blog 
02. 08. 2026 When working with web servers, inspecting log files, or reviewing source code, developers frequently need to locate specific strings or patterns across hundreds of files. The grep command (Global Regular Expression Print) is one of the most powerful and popular command-line utilities in Linux for instant data searching and filtering.
Basic Syntax of the grep Command
The grep command accepts a search pattern, optional flags (switches), and a file or directory path to search within.
| Flag (Option) | Meaning | Usage Description |
|---|---|---|
| -i | Ignore Case | Makes the search case-insensitive (e.g., matching 'error' or 'ERROR'). |
| -r (or -R) | Recursive | Searches recursively through all subdirectories and contained files. |
| -n | Line Number | Displays the line number alongside matching results. |
| -v | Invert Match | Outputs all lines that do not contain the search pattern. |
Practical Search Scenarios in Terminal
The grep tool can be used directly on files or piped (`|`) together with other commands to filter output in real time.
- Searching a single file: Quickly pinpointing a specific error in a log file.
- Recursive project searches: Locating where a specific function or variable is used across PHP code.
- Filtering system outputs: Combining with `ps aux` to find a running process.
Examples of Using grep in Practice
The example below illustrates common usage scenarios in everyday software development and system administration.
# 1. Search for "ERROR" in a log file and show line numbers grep -n "ERROR" /var/log/nginx/error.log # 2. Case-insensitive recursive search across the entire project folder grep -rin "database_connection" /var/www/html/ # 3. Filter running processes – search for MySQL server status ps aux | grep mysql # 4. Count total occurrences of a pattern (-c) grep -c "404" access.log
Power When Combined with Other Tools
The true strength of grep shines when combined with other command-line utilities such as tail (for monitoring live log files, e.g., tail -f access.log | grep "500") or awk for processing and extracting specific data columns.