REDbot

REDbot is lint for HTTP resources.

It checks HTTP resources for feature support and common protocol problems. You can use the public instance on https://redbot.org/, or you can install it locally and use it on the command line, or even self-host your own Web checker.

Contributing to REDbot

Your ideas, questions and other contributions are most welcome. See CONTRIBUTING.md for details.

Setting Up Your Own REDbot

Requirements

REDbot needs:

1. Python 3.6 or greater
2. thor
3. markdown
4. netaddr
5. Jinja2

Once you have Python, you can install the required libraries with:

pip install -r requirements.txt

Installing RED

Unpack the REDbot tarball. The relevant files are:

• bin/redbot_cgi.py - the Web CGI script for running REDbot
• bin/redbot_cli - the command-line interface
• redbot/ - REDbot's Python library files
• redbot/assets/ - REDbot's CSS stylesheet and JavaScript library

To install from source (e.g., if you clone from github):

python setup.py install

installs REDbot's libraries as well as the command-line version as redbot_cli.

Setting up your Web Server

To run REDbot as a CGI script, place redbot_cgi.py where you wish it to be served from by the Web server, and place config.txt in the same directory.

For example, with Apache you can put it in a directory and add these configuration directives (e.g., in `.htaccess, if enabled):

  AddHandler cgi-script .py
  DirectoryIndex redbot_cgi.py

If the directory is the root directory for your server "example.com", this will configure REDbot to be at the URI "http://example.com/".

You can also locate config.txt somewhere else, and indicate its path in an environment variable:

 SetEnv REDBOT_CONFIG /path/to/config.txt

The contents of the assets directory also need to be made available on the server; by default, they're in the 'static' subdirectory of the script's URI. This can be changed using the 'static_root' configuration variable in config.txt.

You should also create the directory referenced by the 'save_dir' configuration variable, and make sure that it's writable to the Web server process. This is where RED stores state files, and you should configure a cron job to regularly clean it. For example:

0 * * * * find /var/state/redbot/ -mmin +360 -exec rm {} ;

Running REDbot as a systemd Service

REDbot can run as a standalone service, managed by systemd. This offers a degree of sandboxing and resource management, as well as process monitoring (including a watchdog function).

To do this, clone the repo to your system and copy extra/redbot.service into the appropriate directory (on most systems, /etc/systemd/system/.)

Modify the file appropriately; this is only a sample. In particular, you will need to adjust the first part of BindReadOnlyPaths to suit the location of the REDbot directory for you.

Then, as root:

systemctl reload-daemon

systemctl enable redbot

systemctl start redbot

By default, REDbot will listen on localhost port 8000. This can be adjusted in config.txt. Running REDbot behind a reverse proxy is recommended, if it is to be exposed to the Internet.

Running REDbot with Docker

If you wish to run REDbot using Docker, get a local copy of the repository, then:

docker build -t redbot .

Start the webserver:

docker run -p 8000:80 redbot

Or, just:

make docker

to run REDbot on port 8000.

Credits

Icons by Font Awesome. REDbot includes code from tippy.js and prettify.js.