fflo/rancher-active-proxy repository overview

If you look for a Kubernetes version : Have a look to Kube Active Proxy⁠

⁠Rancher Active Proxy

Rancher Active Proxy is an all-in-one reverse proxy for Rancher⁠, supporting Letsencrypt out of the box !

Rancher Active Proxy is based on the excellent idea of jwilder/nginx-proxy⁠.

Rancher Active Proxy replaces docker-gen with rancher-gen (a fork of the also excellent janeczku/go-rancher-gen⁠ adding some more functions). Its source now lives in rancher-gen/⁠ in this repository and is compiled as part of the image build.

Rancher Active Proxy use label instead of environmental value.

Images are published to Docker Hub as fflo/rancher-active-proxy⁠ (latest from master, vX.Y.Z from tags) and rebuilt weekly so nginx, certbot and ModSecurity security updates are picked up.

⁠Easy Setup with catalog

Add https://github.com/fflo/rancher-active-proxy.git to your custom catalog list( Rancher > Admin > Settings ).

Then go to catalog and install Rancher Active Proxy !

⁠Usage

Minimal Params To run it:

$ docker run -d -p 80:80 -p 443:443 -v rap-letsencrypt:/etc/letsencrypt fflo/rancher-active-proxy

Then start any containers you want proxied with a label rap.host=subdomain.youdomain.com

$ docker run -l rap.host=foo.bar.com  ...

The containers being proxied must expose⁠ the port to be proxied, either by using the EXPOSE directive in their Dockerfile or by using the --expose flag to docker run or docker create.

Provided your DNS is setup to forward foo.bar.com to the a host running rancher-active-proxy, the request will be routed to a container with the rap.host label set.

⁠Summary of available labels for proxied containers.

Label values are validated before they are written into the nginx configuration: hostnames must be valid DNS names, ports must be numeric, and free-text values (rap.proto, rap.cert_name, rap.client_max_body_size, ...) may not contain whitespace, quotes, ;, {, } or #. A service with an invalid value is skipped (with a comment in the generated config) instead of breaking the proxy for everyone.

⁠Summary of environment variable available for Rancher Active Proxy.

⁠Quick Summary of interesting volume to mount.

⁠Special Attention for standalone containers

Rancher Active Proxy is also able to work for standalone containers on the host it is launched.

There is only one limit to this : You should not use the same host name ( rap.host label ) for a standalone container and for a service.

This feature even enables you to proxy rancher-server, just start it with something like that :

docker run -d --restart=unless-stopped -p 8080:8080 --name=rancher-server -l rap.host=admin.foo.com -l rap.port=8080 -l rap.le_host=admin.foo.com -l [email protected] -l io.rancher.container.pull_image=always rancher/server

In this case admin.foo.com will enable you to acces rancher administration, but it is better to keep port 8080 exposed and use http://foo.com:8080 as the host registration URL.

⁠Let's Encrypt support out of box

Rancher Active Proxy is using certbot from Let's Encrypt in order to automatically get SSL certificates for containers.

In order to enable that feature you need to add rap.le_host label to the container ( you probably want it to be equal to rap.host)

And you should either start Rancher Active Proxy with environment variable DEFAULT_EMAIL or specify rap.le_email as a container label.

If you are developing I recommend to add rap.le_test=true to the container in order to use Let's Encrypt staging environment and to not exceed limits.

⁠SAN certificates

Rancher Active Proxy support SAN certifcates ( one certificate for several domains ).

To create a SAN certificate you need to separate hostnames with ";" ( instead of "," for separate domains)

rap.le_host=admin.foo.com;api.foo.com;mail.foo.com

This will create a single certificate matching : admin.foo.com, api.foo.com, mail.foo.com . The certificate created will be named admin.foo.com but symlink will be create to match all domains.

⁠Multiple Ports

If your container exposes multiple ports, Rancher Active Proxy will use rap.port label, then use the exposed port if there is only one port exposed, or default to DEFAULT_PORT environmental variable ( which is set by default to 80 ). Or you can try your hand at the Advanced rap.host syntax⁠.

⁠Special ByPass for Let's Encrypt

If your container uses its own letsencrypt process to get some certificates Set rap.le_bypass to true to add a location to the http server block to forward /.well-known/acme-certificate/ to upstream through http instead of redirecting it to https

⁠Advanced rap.host syntax

Using the Advanced rap.host syntax you can specify multiple host names to each go to their own backend port. Basically this provides support for rap.host, rap.port, and rap.proto all in one field.

For example, given the following:

rap.host=api.example.com=>http:80,api-admin.example.com=>http:8001,secure.example.com=>https:8443

This would yield 3 different server/upstream configurations...

1. Requests for api.example.com would route to this container's port 80 via http
2. Requests for api-admin.example.com would route to this containers port 8001 via http
3. Requests for secure.example.com would route to this containers port 8443 via https

⁠Multiple Listening Port

If needed you can use Rancher-Active-Proxy to listen for different ports.

docker run -d -p 8081:8081 -p 81:81 fflo/rancher-active-proxy

In this case, you can specify on which port Rancher Active Proxy should listen for a specific hostname :

docker run -d -l rap.host=foo.bar.com -l rap.http_listen_ports="81,8081" -l rap.port="53" containerexposing/port53

In this situation Rancher Active Proxy will listen for request matching rap.host on both port 81 and 8081 of your host and route those request to port 53 of your container.

Likewise, rap.https_listen_ports will work for https requests.

If you are not using port 80 and 443 at all you won't be able to use Let's Encrypt Automatic certificates.

⁠Specific Host Name

Using environmental value SPECIFIC_HOST you can limit Rancher Active Proxy to containers running on a single host.

Just start Rancher Active Proxy like that : docker run -d -p 80:80 -e SPECIFIC_HOST=Hostnameofthehost fflo/rancher-active-proxy

⁠Remove Script

Rancher Active Proxy provides an easy script to revoke/delete a certificate.

Run it inside the running proxy container: docker exec <proxy container> /app/remove admin.foo.com

The argument is the certificate name, i.e. the first host of the rap.le_host label. For a SAN certificate (admin.foo.com;api.foo.com) one call removes the whole certificate and every nginx link that pointed at it. /app/remove '*' deletes all certificates. Names are no longer treated as prefixes/globs.

Do not forget to delete the label on the container before using that script or it will be recreated on next update.

⁠Per-host server configuration

If you want to 100% personalize your server section on a per-rap.host basis, add your server configuration in a file under /etc/nginx/vhost.d The file should use the suffix _server.

For example, if you have a virtual host named app.example.com and you have configured a proxy_cache my-cache in another custom file, you could tell it to use a proxy cache as follows:

$ docker run -d -p 80:80 -p 443:443 -v /path/to/vhost.d:/etc/nginx/vhost.d:ro fflo/rancher-active-proxy

You should therefore have a file app.example.com_server in the /etc/nginx/vhost.d folder that contains the whole server block you want to use :

server {
        server_name app.example.com
        listen 80;
        access_log /var/log/nginx/access.log vhost;

        location / {
                proxy_pass http://app.example.com;
        }
}

If you are using multiple hostnames for a single container (e.g. rap.host=example.com,www.example.com), the virtual host configuration file must exist for each hostname. If you would like to use the same configuration for multiple virtual host names, you can use a symlink.

⁠Per-host server default configuration

If you want most of your virtual hosts to use a default single server block configuration and then override it on a few specific ones, add a /etc/nginx/vhost.d/default_server file. This file will be used on any virtual host which does not have a /etc/nginx/vhost.d/{rap.host}_server file associated with it.

⁠Limit RAP to some containers

If you want a RAP instance to only publish some specific containers/services, you can start the RAP container with environment variable RAP_NAME = example In that situation, all containers to be published by this instance of RAP should have a label rap.rap_name = example If a container should be published by several RAP instances just use a label matching regex like rap.rap_name = internal,external to be published by RAP instance named internal or external

The below part is mostly taken from jwilder/nginx-proxy README⁠ and modified to reflect Rancher Active Proxy

⁠Multiple Hosts

If you need to support multiple virtual hosts for a container, you can separate each entry with commas. For example, foo.bar.com,baz.bar.com,bar.com and each host will be setup the same.

⁠Wildcard Hosts

You can also use wildcards at the beginning and the end of host name, like *.bar.com or foo.bar.*. Or even a regular expression, which can be very useful in conjunction with a wildcard DNS service like xip.io⁠, using ~^foo\.bar\..*\.xip\.io will match foo.bar.127.0.0.1.xip.io, foo.bar.10.0.2.2.xip.io and all other given IPs. More information about this topic can be found in the nginx documentation about server_names⁠.

⁠SSL Backends

If you would like the reverse proxy to connect to your backend using HTTPS instead of HTTP set rap.proto=https on the backend container.

⁠uWSGI Backends

If you would like to connect to uWSGI backend, set rap.proto=uwsgi on the backend container. Your backend container should than listen on a port rather than a socket and expose that port.

⁠Default Host

To set the default host for nginx use the env var DEFAULT_HOST=foo.bar.com for example :

$ docker run -d -p 80:80 -e DEFAULT_HOST=foo.bar.com fflo/rancher-active-proxy

⁠SSL Support

SSL is supported using single host, wildcard and SNI certificates using naming conventions for certificates or optionally specifying a cert name (for SNI) as an environment variable.

To enable SSL:

$ docker run -d -p 80:80 -p 443:443 -v /path/to/certs:/etc/nginx/certs  fflo/rancher-active-proxy

The contents of /path/to/certs should contain the certificates and private keys for any virtual hosts in use. The certificate and keys should be named after the virtual host with a .crt and .key extension. For example, a container with label rap.host=foo.bar.com should have a foo.bar.com.crt and foo.bar.com.key file in the certs directory.

If you are running the container in a virtualized environment (Hyper-V, VirtualBox, etc...), /path/to/certs must exist in that environment or be made accessible to that environment. By default, Docker is not able to mount directories on the host machine to containers running in a virtual machine.

⁠Diffie-Hellman Groups

Let's Encrypt hosts use the RFC 7919 ffdhe2048 group, installed as /etc/letsencrypt/dhparam.pem on first start (replace that file with your own parameters if you prefer; it is never overwritten). For manually provided certificates, the file should be named after the virtual host with a dhparam suffix and .pem extension. For example, a container with rap.host=foo.bar.com should have a foo.bar.com.dhparam.pem file in the certs directory.

⁠Wildcard Certificates

Wildcard certificates and keys should be named after the domain name with a .crt and .key extension. For example rap.host=foo.bar.com would use cert name bar.com.crt and bar.com.key.

⁠SNI

If your certificate(s) supports multiple domain names, you can start a container with rap.cert_name=<name> to identify the certificate to be used. For example, a certificate for *.foo.com and *.bar.com could be named shared.crt and shared.key. A container running with rap.host=foo.bar.com and rap.cert_name=shared will then use this shared cert.

⁠How SSL Support Works

The TLS configuration follows the current Mozilla "intermediate" profile⁠: TLS 1.2 and 1.3 only, AEAD ciphers with forward secrecy, client cipher preference, HTTP/2, session cache without session tickets, and HSTS (tunable per host with rap.hsts). Requests for a host name that is not configured (or a bare IP) are answered with a generated self-signed certificate and a 503, so a tenant's certificate is never presented for a foreign host name. Place your own default.crt/default.key in /etc/nginx/certs to override it. server_tokens is off by default.

The default behavior for the proxy when port 80 and 443 are exposed is as follows:

• If a container has a usable cert, port 80 will redirect to 443 for that container so that HTTPS is always preferred when available.
• If the container does not have a usable cert, a 503 will be returned.

In the latter case the self-signed default certificate is used, so the browser gets a certificate warning followed by a 503 instead of a connection error.

To serve traffic in both SSL and non-SSL modes without redirecting to SSL, you can include the label rap.https_method=noredirect (the default is rap.https_method=redirect). You can also disable the non-SSL site entirely with rap.https_method=nohttp. rap.https_method must be specified on each container for which you want to override the default behavior. If rap.https_method=noredirect is used, Strict Transport Security (HSTS) is disabled to prevent HTTPS users from being redirected by the client. If you cannot get to the HTTP site after changing this setting, your browser has probably cached the HSTS policy and is automatically redirecting you back to HTTPS. You will need to clear your browser's HSTS cache or use an incognito window / different browser.

⁠Basic Authentication Support

In order to be able to secure your virtual host, you have to create a file named as its equivalent rap.host label on directory /etc/nginx/htpasswd/rap.host

$ docker run -d -p 80:80 -p 443:443 \
    -v /path/to/htpasswd:/etc/nginx/htpasswd \
    -v /path/to/certs:/etc/nginx/certs \
    fflo/rancher-active-proxy

You'll need apache2-utils on the machine where you plan to create the htpasswd file. Or you can use an nginx container to create the file ( using OpenSSL as explained in Nginx Readme⁠ )

docker run -it nginx printf "Username_to_use:$(openssl passwd -crypt Password_to_use)\n" >> /path/to/htpasswd/{rap.host}

A default htpasswd can be used to secure all hosts using this proxy. Good for development environments to keep prying eyes out. To use, create the htpasswd file named 'default' here: /etc/nginx/htpasswd/default.

⁠Custom Nginx Configuration

If you need to configure Nginx beyond what is possible using environment variables, you can provide custom configuration files on either a proxy-wide or per-rap.host basis.

⁠Replacing default proxy settings

If you want to replace the default proxy settings for the nginx container, add a configuration file at /etc/nginx/proxy.conf. A file with the default settings would look like this:

# HTTP 1.1 support
proxy_http_version 1.1;
proxy_buffering off;
proxy_set_header Host $http_host;
proxy_set_header Upgrade $http_upgrade;
proxy_set_header Connection $proxy_connection;
proxy_set_header X-Real-IP $remote_addr;
proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
proxy_set_header X-Forwarded-Proto $proxy_x_forwarded_proto;
proxy_set_header X-Forwarded-Port $proxy_x_forwarded_port;

# Mitigate httpoxy attack (see README for details)
proxy_set_header Proxy "";

NOTE: If you provide this file it will replace the defaults; you may want to check the nginx.tmpl file to make sure you have all of the needed options.

NOTE: The default configuration blocks the Proxy HTTP request header from being sent to downstream servers. This prevents attackers from using the so-called httpoxy attack⁠. There is no legitimate reason for a client to send this header, and there are many vulnerable languages / platforms (CVE-2016-5385, CVE-2016-5386, CVE-2016-5387, CVE-2016-5388, CVE-2016-1000109, CVE-2016-1000110, CERT-VU#797896).

⁠Proxy-wide

To add settings on a proxy-wide basis, add your configuration file under /etc/nginx/conf.d using a name ending in .conf.

This can be done in a derived image by creating the file in a RUN command or by COPYing the file into conf.d:

FROM fflo/rancher-active-proxy
RUN { \
      echo 'server_tokens off;'; \
      echo 'client_max_body_size 100m;'; \
    } > /etc/nginx/conf.d/my_proxy.conf

Or it can be done by mounting in your custom configuration in your docker run command:

$ docker run -d -p 80:80 -p 443:443 -v /path/to/my_proxy.conf:/etc/nginx/conf.d/my_proxy.conf:ro fflo/rancher-active-proxy

⁠Per-VIRTUAL_HOST

To add settings on a per-rap.host basis, add your configuration file under /etc/nginx/vhost.d. Unlike in the proxy-wide case, which allows multiple config files with any name ending in .conf, the per-rap.host file must be named exactly after the rap.host.

In order to allow virtual hosts to be dynamically configured as backends are added and removed, it makes the most sense to mount an external directory as /etc/nginx/vhost.d as opposed to using derived images or mounting individual configuration files.

For example, if you have a virtual host named app.example.com, you could provide a custom configuration for that host as follows:

$ docker run -d -p 80:80 -p 443:443 -v /path/to/vhost.d:/etc/nginx/vhost.d:ro fflo/rancher-active-proxy
$ { echo 'server_tokens off;'; echo 'client_max_body_size 100m;'; } > /path/to/vhost.d/app.example.com

If you are using multiple hostnames for a single container (e.g. rap.host=example.com,www.example.com), the virtual host configuration file must exist for each hostname. If you would like to use the same configuration for multiple virtual host names, you can use a symlink:

$ { echo 'server_tokens off;'; echo 'client_max_body_size 100m;'; } > /path/to/vhost.d/www.example.com
$ ln -s /path/to/vhost.d/www.example.com /path/to/vhost.d/ex