After working through this a few days ago I decided to post some detailed instructions to help other get it going. They are available here -
I’m happy to answer questions or accept feedback if anyone decides to use them.
very good tutorial, I’m waiting for further steps because I have a need that I can’t figure out how to solve. I have proxmox with 2 vm running 2 home assistant, in 1 of them I have, as addons, pihole, duckdns, nginx manager, in the other all the other domotic stuff (i.e. switch, lights, automations and so on). I followed your tuto and now I can access to 1 HA dashboard (the one with nginx, so I forwarded 443 and 80 ports to the local ip vm where all is running, in my case 192.168.0.8) digiting mypersonal.duckdns.org (even if I digit http it go automatically to https://mypersonal.duckdns and with no 8123 port in the url, because is nginx that says to use that port). Ok till here is all ok BUT
how can access always fron external to the other vm running HA? It has local ip 192.168.0.42 (I uninstalled duckdns and in configuration.yaml not specified a http base url). In nginx I have 502 bad gateway, it will be usefull to have tips to access all internal local servers such as proxmox dasboard, other HA vm, pihole dashboard, grafana, webcamera to their setup page etc…Thank you in advance.
How can i reset the password? I’ve tried reinstalling but that didn’t work
I ended up not saving my password accidentally on the first go around had to reset as well. I am not sure where the password is stored or how to reset it. However what I noticed is that when you remove the Nginix Proxy Manager it does not remove the database from the MariaDB addon. I am guessing the password is in that database somewhere (I could not find it). What worked for me was to:
- Remove the Nginix proxy manager.
- Install phpMyAdmin from the community store and use it to delete the Nginix Proxy Manger database from MariaDB.
- Reinstall Nginix Proxy Manager and start over.
The main drawback here is that YOU ARE GOING TO LOSE EVERYTHING YOU’VE SET UP in the proxy manager.
If you haven’t set too much up so far it might be worth it though.
2 Likes
I was actually working on a second post but found some mold in the shower which has put me on a little side project. 
I should have the second post up later this week.
Here’s a second post that shows an example of accessing both a service externally and internally only. I hope it helps. https://bobgrayblog.wordpress.com/2020/03/24/using-your-proxy-manager-for-additional-services/
Hi, I had a working addon until this morning, now it does not start, and have this error in LOG
My setup is HASSIIO 107.7 non HASSoS on a NUC (with proxmox). This setting I have since many months now, IO have no idea what is using port 443 (no other addon is using 443)
20-03-31 16:09:43 ERROR (SyncWorker_11) [supervisor.docker] Can’t start addon_a0d7b954_nginxproxymanager: 500 Server Error: Internal Server Error (“driver failed programming external connectivity on endpoint addon_a0d7b954_nginxproxymanager (ebd6f4207b2e06afae2de995cf913a95f33dc0c1f204dafe391ded65f5bb866b): Bind for 0.0.0.0:443 failed: port is already allocated”)
receive also this error not sure is related
20-03-31 16:10:28 WARNING (MainThread) [supervisor.tasks] Watchdog found a problem with CoreDNS plugin!
20-03-31 16:10:28 ERROR (MainThread) [supervisor.tasks] CoreDNS plugin is in fails state / Reset config
20-03-31 16:10:28 INFO (MainThread) [supervisor.dns] Start CoreDNS plugin
20-03-31 16:10:28 INFO (SyncWorker_3) [supervisor.docker.interface] Clean hassio_dns application
20-03-31 16:10:29 INFO (SyncWorker_3) [supervisor.docker.dns] Start DNS homeassistant/amd64-hassio-dns with version 1 - 172.30.32.3
20-03-31 16:10:48 WARNING (MainThread) [supervisor.tasks] Watchdog found a problem with CoreDNS plugin!
20-03-31 16:10:48 ERROR (MainThread) [supervisor.tasks] CoreDNS plugin is in fails state / Reset config
20-03-31 16:10:48 INFO (MainThread) [supervisor.dns] Start CoreDNS plugin
Hi, I had the same issue, please see this GitHub issue: https://github.com/home-assistant/supervisor/issues/1615#event-3180529940
So I have been using this for sometime. Great add-on thanks so much for contributing. I now need to have the http: component in hassion for an integration I’m trying to add, Is there a way to use this as well as enable the http: component?
For some reason the link wasn’t working, but I figured it out: Part 1 and Part 2
Thank you so much for taking the time to write these instructions. I’m not familiar with dns- and proxy server just enough to make it work myself, but thanks to your very clear instructions I managed to make it work!
And more importantly: I think I sort of understand how this all works together! 
1 Like
I’m glad it helped and thanks for letting me know! 
Hi,
I’m trying to run this add on. I have installed the mariadb ad on and started that, with a strong unique password but otherwise unchanged.
I now click start on this add on, but it simply doesn’t start (start button turns red).
I all ready have https (certificate generated by my qnap server) and duckdns (home assistant add on) set up and running fine, I just want to use this to forward a few subfolders to a few additional services I already have set up (but which are not currently externally accessible).
Any ideas why it wouldn’t start up?
1 Like
PS - even with trace level logs, nothing is output in the add-on logs.
PPS - I am running latest version of Hassio on a raspberry pi 4, with latest (non Beta) version of home assistant and supervisor.
1 Like
How to use this add on if i don’t wanna forward port 443 from my router to my raspberry? I would like to open for example port 11234 to my raspberry. What do i have to do to get the add on working with this port?
Trying to add a new subdomain but I receive this error, anyone knows why?
4/15/2020] [2:59:50 PM] [Express ] › 
warning Command failed: /usr/bin/certbot certonly --non-interactive --config “/etc/letsencrypt.ini” --cert-name “npm-8” --agree-tos --email “[email protected]” --preferred-challenges “dns,http” --webroot --domains “xxxx.duckdns.org”
Saving debug log to /data/logs/letsencrypt/letsencrypt.log
Plugins selected: Authenticator webroot, Installer None
Obtaining a new certificate
Performing the following challenges:
http-01 challenge for xxxx.duckdns.org
Using the webroot path /data/letsencrypt-acme-challenge for all unmatched domains.
Waiting for verification…
Challenge failed for domain xxxx.duckdns.org
http-01 challenge for xxxx.duckdns.org
Cleaning up challenges
Some challenges have failed.
Is it possible to have HA as a sub folder?
Like subdomain.mydomain.com/ha while having another site on subdomain.mydomain.com configured on NPM.
I tried adding a custom path and using rewrite /ha/(.*) /$1 break; as suggested in various posts but i have a malformatted page as result ( a blank page with HACS text ).
Is there a solution for this?
Thanks
Might work but not supported (go through the above posts): https://github.com/home-assistant/architecture/issues/156#issuecomment-474528786
Ah!
Didn’t know that, I was thinking it could it be something wrong with the addon or some weird config to use.
Thanks for the info 
Hi, all! I’am trying enable nginx proxy manager addon without any domain names, using IPs only. Why it more difficult then with names. I have wireguard addon with own subnet? And I want any reqwest from its subnet redirects to another IPs. Can anybody little help me? M?)