[Support] Nginx Proxy Manager (NPM) Official


Recommended Posts

1 hour ago, mgutt said:

Please follow the debug 5xx error part on the first page.


Based on that debug, this is my issue...

This time you need to enter "http://[8888:7777:6666:5555]" (the square brackets are important and again do not use "https://"!).

That did not work.  Having said that, I'm still confused.  I double checked all my port forwarding stuff, and everything looks correct.

I don't understand why/how it would be different on a Ubiquiti UDM Pro vs a Netgear router.

Here are the Port forward rules I have setup in my system.  I even checked under firewall rules, and it looks ok.

 

2022-12-05_14-58-51.jpg

2022-12-05_14-56-57.jpg

2022-12-05_14-57-12.jpg

2022-12-05_14-57-35.jpg

2022-12-05_14-57-53.jpg

2022-12-05_10-31-31.jpg

Link to comment
8 minutes ago, coltonc18 said:

This time you need to enter "http://[8888:7777:6666:5555]" (the square brackets are important and again do not use "https://"!).

This explanation is for IPv6. Does your DDNS domain have an IPv6 record / are you really using IPv6?! Most people use only IPv4.

Link to comment
1 hour ago, mgutt said:

This explanation is for IPv6. Does your DDNS domain have an IPv6 record / are you really using IPv6?! Most people use only IPv4.

 

There goes my reading comprehension....for some reason I was thinking the next part was not....

Anyways after reading everything again, and going through it all.

My issue appears to be at the curl, when I go to test the curl nothing happens.  I can open overseer and it does work at this IP, so I know that is not the issue, reading down below that I wasn't sure what action to take next. 

I tried to change the Bridge network on Overseer to Custom Br0 to match this, but then Overseer wouldn't work, and I tried to do the same on binhex-nginx, and while that one allowed the change I still couldn't access the page attached to it.  

When I tried to change the network on nginx to bridge it just didn't work.  The strange thing is all of these worked as they are right now, the only thing I changed was my router this weekend, and I have been working on getting everything set back up since then. 

Thank you for your patience with me! 

 

2022-12-05_15-33-17.jpg.39b2c56999a319923eb269484ad24b3d.jpg

Edited by coltonc18
Link to comment
1 hour ago, coltonc18 said:

My issue appears to be at the curl, when I go to test the curl nothing happens

Looks good to me. The response is possibly empty because it redirects. Use -LsS as curl options and curl follows the redirect and you will see the html content if the page.

 

If this works, than NPM reaches the target container.

 

Does your new router support NAT Loopback / Hairpinning? This must be enabled to reach a domain with the same public IP as your internet connection.

 

Link to comment
2 hours ago, mgutt said:

Looks good to me. The response is possibly empty because it redirects. Use -LsS as curl options and curl follows the redirect and you will see the html content if the page.

 

If this works, than NPM reaches the target container.

 

Does your new router support NAT Loopback / Hairpinning? This must be enabled to reach a domain with the same public IP as your internet connection.

 

Doing the -LsS it redirects to where it should....when looking up the Ubiquity UDM Pro it says it supports loopback/hairpinning....IT should considering its a high end commerical system, guess i'll have to ask them about it....

Link to comment
6 hours ago, mgutt said:

So NPM is able to reach the container. And you reached NPN through the public IPv4?

I can reach overseer through my local IP that I use, and when I used - LsS I saw Nginx was able to access overseer based on the data that came back. But when I try to access overseer through the proxy domain I created in and put in Nginx I get the 5xx error. 

 

I spoke to some people on ubiquiti forums last night, and this system does have NAT loop back, so I'm a bit confused at this point. 

Link to comment

Slight update....I deleted the container completed and the app folder and started from scratch, and cant get a SSL certificate, but when I put my public IP out there I do get the Nginx home page, unfortunately it just means the same thing, somehow it can't reach the containers.  

The error I'm seeing in the logs there is 

Detail: 111.11.111.111: Fetching http://notifiarr.duckdns.org/.well-known/acme-challenge/abErDJIBoJR6U8sWtF0qofN4BmPSTjN6hwaFcmxWRH4: Timeout during connect (likely firewall problem)

I'm at a loss as to what the firewall problems can be though.  As the port forwarding seems to be working, or I wouldn't be getting the welcome screen. 

Link to comment
1 hour ago, coltonc18 said:

public IP out there I do get the Nginx home page

 

1 hour ago, coltonc18 said:

Timeout during connect (likely firewall problem)

 

If you can reach NPM through your public IP, but Let's Encrypt can't reach NPM through your domain, then there are two possible reasons:

- your domain does not use your public IP

- your public IP isn't public

 

 

 

Link to comment
44 minutes ago, mgutt said:

 

 

If you can reach NPM through your public IP, but Let's Encrypt can't reach NPM through your domain, then there are two possible reasons:

- your domain does not use your public IP

- your public IP isn't public

 

 

 

I'm at my daughters gymnastics right now, so on my cell, and just tried the public IP and it worked.... I tried adding :5055 on the end of it, and it didn't work. I'm not sure if that says something or not.

 

So since the second was correct, my public IP is public, how do I go about researching the 1st issue more?

 

These are duckdns domains, is this an issue with that some how? Or something else? 

 

Thanks again! 

Link to comment
2 hours ago, coltonc18 said:

I'm at my daughters gymnastics right now, so on my cell, and just tried the public IP and it worked.... I tried adding :5055 on the end of it, and it didn't work. I'm not sure if that says something or not.

 

So since the second was correct, my public IP is public, how do I go about researching the 1st issue more?

 

These are duckdns domains, is this an issue with that some how? Or something else? 

 

Thanks again! 

Wow!! After the last message when I got home and put the girls to bed I started looking at this, and looked at duckdns......I noticed there was a current ip listed there....it listed my old IP, and not my new one.  I updated the IP to show my new one, and bam it worked!! I had no idea something like that was needed, I never did that the first time I did this.

Thanks again for all your time you spent trying to help me figure this out! I truly appreciate it!!

Link to comment
5 hours ago, Kilrah said:

not having port 80 forwarded

As he was able to see the NPM page through the public IP, this couldn't be the reason.

 

3 hours ago, coltonc18 said:

  I updated the IP

Aren't you using the duckdns container? It updates the public IP of your domain(s) automatically.

Link to comment

Hi Team, hoping for some help.

Have NPM up and running without issues but struggling to get custom location working. inteded use case is to use existing domain with /photos/ to redirect to different IP within network but receiving 502 error. Screen shot of custom location configuration below.


Have gotten new SSL certificate via LetsEncrypt.

 

Any suggestions?

image.png.0add67c917f9562535db42c0f88e06f2.png

Edited by Rutj87
Link to comment
1 hour ago, mgutt said:

Follow the debug 5xx steps on the first page.

Thanks for the rapid response @mgutt

 

Have completed 5XX debug steps and confirm

1. NPM container working

2. NPM online and existing proxy host working

3. Target container 192.168.86.244:80 working and accessible

4. NPM can reach target container

 

Let me know what other information you need :) 

 

I am not getting a 502 error anymore but get a blank page and all that loads is the tab heading "Librephotos" which is what it should be opening.

 

TIA

Edited by Rutj87
Link to comment
3 hours ago, Rutj87 said:

all that loads is the tab heading "Librephotos" which is what it should be opening.

Open the browsers debug tools and check the network monitor (reload page to see all requests).

 

I guess it's something regarding the subdirectory. Which means the website can not run inside a subdir or the subdir must be added somewhere in a config file of the website.

Link to comment
8 hours ago, mgutt said:

As he was able to see the NPM page through the public IP, this couldn't be the reason.

 

Aren't you using the duckdns container? It updates the public IP of your domain(s) automatically.

Thanks for the reminder...I checked it out and I had never updated it when I created these domains.  It only had my first domain on there, which ironically I don't use anymore.  I have added them all to it now, so this shouldn't happen again.  Thank you again for your time and patience with me. 

Link to comment
9 hours ago, mgutt said:

Open the browsers debug tools and check the network monitor (reload page to see all requests).

 

I guess it's something regarding the subdirectory. Which means the website can not run inside a subdir or the subdir must be added somewhere in a config file of the website.

Thanks @mgutt

 

Hopefully this is the info need and have also current config from NPM

 

image.png.e94b1265da6a6e3a9a7e9b91453fe8b6.png

image.png

Link to comment
1 minute ago, mgutt said:

Check the network monitor, not sources. And don't forget to reload the page. The network monitor must be open for the complete loading process.

Apologies, hopefully right this time.

 

1st image is trying to run via custom location and 2nd is normal loading directly with target ip

 

 

image.png

image.png

Link to comment
44 minutes ago, Rutj87 said:

1st image

As you can see. It can't load the JS and CSS file. You can click on the entry and it will show you the path. It think it tries to find the JS file in the root directory if the domain and as you are in the subdir, this path does not exist.

 

Now you need to find out if this container / software allows to set the "base url" from / to /subdir. So this has nothing which can be solved through NPM.

Link to comment
7 hours ago, mgutt said:

As you can see. It can't load the JS and CSS file. You can click on the entry and it will show you the path. It think it tries to find the JS file in the root directory if the domain and as you are in the subdir, this path does not exist.

 

Now you need to find out if this container / software allows to set the "base url" from / to /subdir. So this has nothing which can be solved through NPM.

@mgutt - it seems its now reading from base url correctly but giving style error. Is this something with the config of NPM and the mime types enabled?

 

image.thumb.png.7658a52a137a6537a7c7391f752d21ca.png

Link to comment

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Restore formatting

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.