Right now, there are three possible authentication methods. Each one of them has its own capabilities and specification. If you are interested in contributing with one more authentication method, please check the guidelines.
We call it JSON Authentication but it is just the default authentication method and the one that is provided by default if you don't make any changes. It is set by default, but if you've made changes before you can revert to using JSON auth:
filebrowser config set --auth.method=json
This method can also be extended with reCAPTCHA verification during login:
filebrowser config set --auth.method=json \
--recaptcha.key site-key \
By default, we use Google's reCAPTCHA service. If you live in China, or want to use other provider, you can change the host with the following command:
filebrowser config set --recaptcha.host https://recaptcha.net
https://recaptcha.netis any provider you want.
Note that you always need to set the
--auth.methodflag when changing authentication configurations and that it will completely overwrite your current settings. This is a known issue.
If you have a reverse proxy you want to use to login your users, you do it via our
proxyauthentication method. To configure this method, your proxy must send an HTTP header containing the username of the logged in user:
filebrowser config set --auth.method=proxy --auth.header=X-My-Header
X-My-Headeris the HTTP header provided by your proxy with the username.
File Browser will blindly trust the provided header. If the proxy can be bypassed, an attacker could simply attach the header and get admin access.
We also provide a no authentication mechanism for users that want to use File Browser privately such in a home network. By setting this authentication method, the user with id 1 will be used as the default users. Creating more users won't have any effect.
filebrowser config set --auth.method=noauth