indgy / app-jail

A Bastille template to create an empty jail

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

App Jail

A small script to create the mimimal possible BastilleBSD jail to run a binary application supervised by daemon.

Why no Bastillefile?

Bastillefiles are a great way to provision jails, however Bastille will not apply templates to a stopped jail and an empty jail cannot be started, so we have to bootstrap it the old fashioned way.

Usage

Call this script with the full path to your jail and the IP address you want to use.

./sh-jail jail_name ip.ad.dr.ess

This uses the default bastille0 shared interface.

You may specify an alternative as the final parameter.

./sh-jail jail_name ip.ad.dr.ess bridge0

The script will create an empty Bastille jail and set it up with the bare minimum structure and binaries to run a binary supervised by the daemon process.

Install your app

Copy your app to the jailed /bin folder, you may need copy any supporting files or folders into the jail.

Update the exec.start parameter in jail.conf to point to your app binary.

Start the jail

Start the jail and your app will start:

bastille start jail_name

Check that your app is listening:

nc -z ip.ad.dr.ess port

If all went well you should see a connection succeeded message, now you need to tell your host or reverse proxy to forward traffic to your jail.

Note: When restarting the jail you may see an error saying the IP is in use, you must remove the jail ip manually:

ifconfig bastille0 delete ip.ad.dr.ess

For some reason this is not run by exec.release in jail.conf.

About

A Bastille template to create an empty jail

License:MIT License


Languages

Language:Shell 100.0%