Skip to content

Latest commit

 

History

History
106 lines (69 loc) · 4.75 KB

README.md

File metadata and controls

106 lines (69 loc) · 4.75 KB

Async DNS Brute

A Python 3.5+ tool that uses asyncio to brute force domain names asynchronously.

aiodnsbrute screenshot

Speed

It's fast. Benchmarks on small VPS hosts put around 100k DNS resoultions at 1.5-2mins. An amazon M3 box was used to make 1 mil requests in just over 3 minutes. Your mileage may vary. It's probably best to avoid using Google's resolvers if you're purely interested in speed.

DISCLAIMER

  • Your ISP's and home router's DNS servers probably suck. Stick to a VPS with fast resolvers (or set up your own) if you're after speed.
  • WARNING This tool is capable of sending LARGE amounts of DNS traffic. I am not repsonsible if you DoS someone's DNS servers.

Installation

$ pip install aiodnsbrute

Note: using a virtualenv is highly recommended.

Alternate install

Alternately you can install the usual way:

$ git clone https://github.com/blark/aiodnsbrute.git
$ cd aiodnsbrute
$ python setup.py install .

Usage

Get help:

$ aiodnsbrute --help

Usage: cli.py [OPTIONS] DOMAIN

  aiodnsbrute is a command line tool for brute forcing domain names
  utilizing Python's asyncio module.

  credit: blark (@markbaseggio)

Options:
  -w, --wordlist TEXT           Wordlist to use for brute force.
  -t, --max-tasks INTEGER       Maximum number of tasks to run asynchronosly.
  -r, --resolver-file FILENAME  A text file containing a list of DNS resolvers
                                to use, one per line, comments start with #.
                                Default: use system resolvers
  -v, --verbosity               Increase output verbosity
  -o, --output [csv|json|off]   Output results to DOMAIN.csv/json (extension
                                automatically appended when not using -f).
  -f, --outfile FILENAME        Output filename. Use '-f -' to send file
                                output to stdout overriding normal output.
  --query / --gethostbyname     DNS lookup type to use query (default) should
                                be faster, but won't return CNAME information.
  --wildcard / --no-wildcard    Wildcard detection, enabled by default
  --verify / --no-verify        Verify domain name is sane before beginning,
                                enabled by default
  --version                     Show the version and exit.
  --help                        Show this message and exit.

Examples

Run a brute force with some custom options:

$ aiodnsbrute -w wordlist.txt -vv -t 1024 domain.com

Run a brute force, supppess normal output and send only JSON to stdout:

$ aiodnbrute -f - -o json domain.com

...for an advanced pattern, use custom resovers and pipe output into the awesome jq:

$ aiodnsbrute -r resolvers.txt -f - -o json google.com | jq '.[] | select(.ip[] | startswith("172."))'

Wildcard detection enabled by default (--no-wildcard turns it off):

$ aiodnsbrute foo.com

[*] Brute forcing foo.com with a maximum of 512 concurrent tasks...
[*] Using recursive DNS with the following servers: ['50.116.53.5', '50.116.58.5', '50.116.61.5']
[!] Wildcard response detected, ignoring answers containing ['23.23.86.44']
[*] Wordlist loaded, proceeding with 1000 DNS requests
[+] www.foo.com                         52.73.176.251, 52.4.225.20
100%|██████████████████████████████████████████████████████████████████████████████| 1000/1000 [00:05<00:00, 140.18records/s]
[*] Completed, 1 subdomains found

NEW use gethostbyname (detects CNAMEs which can be handy for potential subdomain takeover detection)

$ aiodnsbrute --gethostbyname domain.com

Supply a list of resolvers from file (ignoring blank lines and starting with #), specify -r - to read list from stdin.

$ aiodnsbrute -r resolvers.txt domain.com

Thanks

  • Wordlists are from bitquark's dnspop repo (except the 10 mil entry one which I created using his tool).
  • Click for making CLI apps so easy.
  • tqdm powers the pretty progress bar!
  • aiodns for providing the Python async interface to pycares which makes this all possible!

Notes

  • You might want to do a ulimit -n to see how many open files are allowed. You can also increase that number using the same command, i.e. ulimit -n <2048>