NZBmegasearcH is a great program. It aggregates all the sites I regularly visit to find an NZB to just one place. CouchPotato and SickBeard can also retrieve results from it instead of querying multiple locations for updates. One thing it doesn't do well is being compatible with reverse proxy which 0byte admits. But that doesn't mean its impossible!
It took me two days to figure this out. A lot of 'sudo service nginx restart' and 'CTRL + SHIFT + R' presses. To sum up experiences:
- NGINX cannot proxy it
- Apache can proxy it
- NGINX can proxy Apache
- NGINX can proxy SSL Apache
Did you get that? This guide is for users running NGINX as their proxy instead of Apache. If you're using Apache, jump to my configuration file for your answer!
Let's run through how my setup works.
My Setup
In my environment the proxy server is a different host from NZBmegasearcH. My proxy server connects to all my Usenet applications which are all on different virtualised hosts. Here's a little picture to help you out with that:
Only my proxy server is exposed to the Internet. It connects to other hosts when accessing applications.:
Apache Proxy Install
1. I recommend you pull my fork of NZBmegasearcH I've fixed a few errors in code such as SSL problems, favicons not showing in proxy and dropdown arrows not displaying.
2. Install Apache 2 on to the NZBmegasearcHhost. DigitalOcean have a terrific guide on installing Apache with SSL configuration.
3. Let's try to get NZBmegasearcH up and running on its own local proxy using Apache. I found that NGINX couldn't do this for whatever reason. You just want it to show any content for now, it doesn't have to be perfect and styled. This is my Apache configuration at /etc/apache2/sites-enabled/000-default.conf. Thanks to goatysix for this configuration.
<VirtualHost *:80>
ServerAdmin webmaster@localhost
DocumentRoot /var/www/html
ErrorLog ${APACHE_LOG_DIR}/error.log
CustomLog ${APACHE_LOG_DIR}/access.log combined
</VirtualHost>
<VirtualHost *:443>
SSLEngine on
SSLCertificateFile /etc/apache2/ssl/server.crt
SSLCertificateKeyFile /etc/apache2/ssl/server.key
</VirtualHost>
<Location /nzbmegasearch>
ProxyPass http://localhost:5000
ProxyPassReverse http://localhost:5000
</Location>
<Location /nzbmegasearch/api>
Satisfy Any
Allow from all
</Location>
<Location /nzbmegasearch/warp>
Satisfy Any
Allow from all
</Location>Apache is proxy passing NZBmegasearcH on localhost:5000 over to /nzbmegasearch. Save and restart Apache then browse over to http://apache2host/nzbmegasearch/ to give it a try. Don't forget the trailing slash (/) or it may not load the stylesheets!
If it works, let's move onto the next step.
NZBmegasearcH Configuration
Within NZBmegasearcH there is an option to set the Reverse proxy directory located in the 'Connectivity' section. I found that you have to put the full URL for this to work.
This means the URL you will be accessing it from including the protocol, domain and path. So if you were accessing it over HTTPS, your domain is mydomain.com and the path will be /search then you will have to put in https://mydomain.com/search. Save and restart NZBmegasearcH after you have done this.
After it restarts, when attempting a search it will redirect you to your new URL + whatever you had searched for. This is a good sign, even though it doesn't work. Now we'll get NGINX to proxy Apache to retrieve the right location.
NGINX Configuration
More or less a simple 'proxy_pass' to the Apache URL you accessed earlier.
location /nzbmegasearch {
#put your apache URL you used earlier to access NZBmegasearcH
proxy_pass https://apache2host/nzbmegasearch/;
}Save and restart NGINX once you're done.
Try It Out
If all goes well, simply browsing over to the reverse proxy URL (e.g. the one we specified earlier https://mydomain.com/search/) will bring up NZBmegasearcH! Don't forget the trailing slash (/) or it may not apply the stylesheets. Try performing a search and sending it to SABnzbd. The Apache locations for 'api' and 'wrap' should have you covered.
I found an issue if you use HTTP authentication on Apache or NGINX. SABnzbd would be get notified to download a file from '/nzbmegasearch/warp' but won't be able to access it due to the authentication required. A simple workaround is to just allow the path through.
location /nzbmegasearch {
include auth.conf;
proxy_pass https://apache2host/nzbmegasearch/;
}
location /nzbmegasearch/warp {
#disable authentication include auth.conf;
proxy_pass https://apache2host/nzbmegasearch/warp;
}Let's Wrap It Up
What we've done is have NGINX proxy access a Apache Proxy accessing NZBmegasearcH. There is little benefit in enabling HTTPS on NZBmegasearcH as it is on the localhost with Apache. I couldn't get that to work though, but if does for you let me know! If you were running Apache as your Proxy Server then this whole tutorial was moot.
Luckily the other Usenet programs out there aren't as complicated to reverse proxy and generally you will find a lot of configurations posted online for both NGINX and Apache, maybe even Lighttpd and IIS?