Quote from joshua.michael.anderson on May 19, 2018, 10:47 pm

I'm a bit baffled and unsure if this is an apache2 setup problem or if something else is interrupting the routing from nginx.

A little history so you can understand my setup.  I am migrating an old desktop ubuntu machine running 16.04, and providing me multiple bits of functionality (dns, ldap, apache for reverse proxy).  I have moved the DNS to an antlet and it is working fine and it is now serving my internal network.  I have moved apache2 to an antlet and it is working fine for both my internal network's naming conventions, and my public facing DNS name (jma.is-a-geek.com) which I'm using DynDns to keep it updated.  On all of the machines in my house the networking and routing and reverse proxy works great.  I have the apache2 antlet with a public IP, and I have my router forward traffic forward to that external IP.   I can call the index.html landing page from both in my network and off my network (using my phone and dropping wifi), and I get the index.html with the word hello in it.

 

For the life of me I can't access any of the context paths that I have with a proxy on them from my phone when not on the network.  I monitor the apache2 logs and if I go to the index.html, it will log the access log and everything.  If I go to any of the sub paths, the access log isn't even triggered.  Switch my phone back to the network and everything is working fine and I can access everything.

Could something about the ngix be blocking the reverse proxy?  Any linux buffs out there that have some tips on figuring out where this thing is dying in the routing?

Updated:

So I had already setup {subdomain}.antsle.us for myself, and I noticed that it had generated letsencrypt for its certificate.  I looked at the access by domain name setup and copied the similar configuration files that were already there for the antsle sub domain (http://docs.antsle.com/accdomain/).  I also stopped port forwarding from my router to my apache antlet and sent both HTTP and HTTPS to my antsle, copied the certificates from my antlet to the anstle, and it started working.

So here is what I have changed:

Previous setup

router:443 -> antlet (apache2 reverse proxy) -> other antlets / servers in my private network

Current Setup

router:80/443 -> antsle (nginx proxy) -> antlet (apache2 reverse proxy) -> other antlets / servers in my private network

I am fine with this setup, but it seems to be a little much to have to proxy twice to pull this off.  If people think there is a way to bypass the stop at the antsle nginx, I would love some tips on how to do it.

Also, I had the letsencrypt setup to auto renew on my antlet, but I can't figure out how to do that on the antsle itself.  In fact I can't figure out what the OS is to try to install any packages from letsencrypt.  If anyone has some documentation on what needs to be done for the antsle itself to be able to auto renew certificates via letsencrypt please point me in that direction.