sqlfuzz
sqlfuzz copied to clipboard
Simple SQL table fuzzing
SQLfuzz
Load random data into SQL tables for testing purposes. The tool can get the layout of the SQL table and fill it up with random data.
- Installation
- Usage
- Flags
- Package usage
Installation
MacOS
wget https://github.com/PumpkinSeed/sqlfuzz/releases/download/{RELEASE}/sqlfuzz_darwin_amd64 -O /usr/local/bin/sqlfuzz
chmod +x /usr/local/bin/sqlfuzz
Linux
# amd64 build
wget https://github.com/PumpkinSeed/sqlfuzz/releases/download/{RELEASE}/sqlfuzz_linux_amd64 -O /usr/local/bin/sqlfuzz
chmod +x /usr/local/bin/sqlfuzz
# arm64 build
wget https://github.com/PumpkinSeed/sqlfuzz/releases/download/{RELEASE}/sqlfuzz_linux_arm64 -O /usr/local/bin/sqlfuzz
chmod +x /usr/local/bin/sqlfuzz
Windows
You can download the Windows build here
Build from source
wget https://github.com/PumpkinSeed/sqlfuzz/archive/{RELEASE}.zip
# unzip
# cd into dir
go install main.go
Usage
# MySQL
sqlfuzz -u username -p password -d database -h 127.0.0.1 -t table -n 100000 -w 100
# Postgres
sqlfuzz -u username -p password -d database -h 127.0.0.1 -t table -n 100000 -w 100 -P 5432 -D postgres
Flags
-
u
: User for database connection -
p
: Password for database connection -
d
: Database name for database connection -
h
: Host for database connection -
P
: Port for database connection -
D
: Driver for database connection (supported:mysql
,postgres
) -
t
: Table for fuzzing -
n
: Number of rows to fuzz -
w
: Concurrent workers to work on fuzzing -
s
: Seed value for reproducibility of data
Package usage
TODO: Write package