mattie112 Posted July 29, 2021 Share Posted July 29, 2021 5 hours ago, XisoP said: This might have been asked before. In that case: sorry 😅 I'm running NPM on my unraid server. I'm also running a VM which is serving my access control to my house. This platform is web-based. Is there a way to point <sub.domain.ext> to a site that is not running on my custom docker network? It is http traffic on a fixed IP for the VM. Tried some stuff allready, pointing to IP or hostname, pushing buttons and sliding sliders but all I get is errors. Yes Simpy forward to ip.of.your.vm:portofyourapp does not need to be on the same host 1 Quote Link to comment
XisoP Posted July 29, 2021 Share Posted July 29, 2021 5 hours ago, mattie112 said: Yes Simpy forward to ip.of.your.vm:portofyourapp does not need to be on the same host Figured it out. Turns out that a MAC adress changes on a VM somehow. Had everything configured to look for IP 24 which whas bound by MAC adress. When the MAC doesn't exist in my scope, the machine is assigned an adress from the DHCP pool 😩 Kicked in a static IP in my VM and all is working like a charm. Thanks for your reply Quote Link to comment
Linguafoeda Posted July 30, 2021 Share Posted July 30, 2021 On 7/29/2021 at 9:51 AM, mattie112 said: @Linguafoeda 502 is a problem between NPM and the endpoint. Usally you let NPM handle SSL so: internet -> SSL -> yourdomain -> non-SSL -> your-internal-service (like plex, apache, bitwarden, whatever) Can you share some screenshots? I find it hard to understand exactly what your setup is. I see. I figured it out, had to use http scheme in Nginx Proxy host setup. Also setup wildcard SSL certificate with DNS challenge to get "tautulli.mydomain.duckdns.org" working Quote Link to comment
plantsandbinary Posted August 12, 2021 Share Posted August 12, 2021 (edited) Everything stopped working after a reboot. All I get in the logs are this: [8/12/2021] [1:17:41 PM] [Migrate ] › ℹ info Current database version: none [8/12/2021] [1:17:41 PM] [Setup ] › ℹ info Logrotate Timer initialized [8/12/2021] [1:17:41 PM] [Setup ] › ℹ info Logrotate completed. [8/12/2021] [1:17:41 PM] [IP Ranges] › ℹ info Fetching IP Ranges from online services... [8/12/2021] [1:17:41 PM] [IP Ranges] › ℹ info Fetching https://ip-ranges.amazonaws.com/ip-ranges.json [8/12/2021] [1:17:41 PM] [IP Ranges] › ℹ info Fetching https://www.cloudflare.com/ips-v4 [8/12/2021] [1:17:41 PM] [IP Ranges] › ℹ info Fetching https://www.cloudflare.com/ips-v6 [8/12/2021] [1:17:41 PM] [SSL ] › ℹ info Let's Encrypt Renewal Timer initialized [8/12/2021] [1:17:41 PM] [SSL ] › ℹ info Renewing SSL certs close to expiry... [8/12/2021] [1:17:41 PM] [IP Ranges] › ℹ info IP Ranges Renewal Timer initialized [8/12/2021] [1:17:41 PM] [Global ] › ℹ info Backend PID 2679 listening on port 3000 ... [8/12/2021] [1:17:43 PM] [Nginx ] › ℹ info Reloading Nginx [8/12/2021] [1:17:43 PM] [SSL ] › ℹ info Renew Complete `QueryBuilder#allowEager` method is deprecated. You should use `allowGraph` instead. `allowEager` method will be removed in 3.0 `QueryBuilder#eager` method is deprecated. You should use the `withGraphFetched` method instead. `eager` method will be removed in 3.0 QueryBuilder#omit is deprecated. This method will be removed in version 3.0 Duplicate relation "access_list" in a relation expression. You should use "a.[b, c]" instead of "[a.b, a.c]". This will cause an error in objection 2.0 [8/12/2021] [1:18:59 PM] [Nginx ] › ℹ info Reloading Nginx [8/12/2021] [1:19:01 PM] [Nginx ] › ℹ info Reloading Nginx [8/12/2021] [1:19:02 PM] [Nginx ] › ℹ info Reloading Nginx [8/12/2021] [1:19:04 PM] [Nginx ] › ℹ info Reloading Nginx [8/12/2021] [1:19:05 PM] [Nginx ] › ℹ info Reloading Nginx [8/12/2021] [1:19:06 PM] [Nginx ] › ℹ info Reloading Nginx [8/12/2021] [1:19:08 PM] [Nginx ] › ℹ info Reloading Nginx [8/12/2021] [1:19:09 PM] [Nginx ] › ℹ info Reloading Nginx [8/12/2021] [1:19:10 PM] [Nginx ] › ℹ info Reloading Nginx [8/12/2021] [1:19:11 PM] [Nginx ] › ℹ info Reloading Nginx [8/12/2021] [1:19:12 PM] [Nginx ] › ℹ info Reloading Nginx [8/12/2021] [1:19:13 PM] [Nginx ] › ℹ info Reloading Nginx [8/12/2021] [1:20:50 PM] [Nginx ] › ℹ info Reloading Nginx [8/12/2021] [1:20:57 PM] [Nginx ] › ℹ info Reloading Nginx [8/12/2021] [1:21:02 PM] [Nginx ] › ℹ info Reloading Nginx [8/12/2021] [1:21:07 PM] [Nginx ] › ℹ info Reloading Nginx [8/12/2021] [1:21:14 PM] [Nginx ] › ℹ info Reloading Nginx [8/12/2021] [1:21:18 PM] [Nginx ] › ℹ info Reloading Nginx [8/12/2021] [1:21:23 PM] [Nginx ] › ℹ info Reloading Nginx [8/12/2021] [1:21:29 PM] [Nginx ] › ℹ info Reloading Nginx [8/12/2021] [1:22:37 PM] [Nginx ] › ℹ info Reloading Nginx [8/12/2021] [1:22:44 PM] [Nginx ] › ℹ info Reloading Nginx [8/12/2021] [1:22:52 PM] [Nginx ] › ℹ info Reloading Nginx [8/12/2021] [1:22:58 PM] [Nginx ] › ℹ info Reloading Nginx [8/12/2021] [1:23:03 PM] [Nginx ] › ℹ info Reloading Nginx [8/12/2021] [1:23:09 PM] [Nginx ] › ℹ info Reloading Nginx [8/12/2021] [1:23:17 PM] [Nginx ] › ℹ info Reloading Nginx [8/12/2021] [1:23:24 PM] [Nginx ] › ℹ info Reloading Nginx [8/12/2021] [1:23:30 PM] [Nginx ] › ℹ info Reloading Nginx [8/12/2021] [1:23:34 PM] [Nginx ] › ℹ info Reloading Nginx [8/12/2021] [1:23:39 PM] [Nginx ] › ℹ info Reloading Nginx [8/12/2021] [1:23:45 PM] [Nginx ] › ℹ info Reloading Nginx [8/12/2021] [1:23:49 PM] [Nginx ] › ℹ info Reloading Nginx [8/12/2021] [1:23:54 PM] [Nginx ] › ℹ info Reloading Nginx [8/12/2021] [1:24:00 PM] [Nginx ] › ℹ info Reloading Nginx [8/12/2021] [1:24:05 PM] [Nginx ] › ℹ info Reloading Nginx Cloudfare says it's working but all of my sites are giving a host error. Edited August 12, 2021 by plantsandbinary Quote Link to comment
darrenyorston Posted August 13, 2021 Share Posted August 13, 2021 (edited) Im having difficulty getting Cloudflare, NGINX and any Docker container working. I have working Docker containers that I want to proxy. nextcloud being one. I have the Cloudfare DNS docker container functioning and the appropriate A record for my domain name is showing on Cloudflare. I have created a CNAME for nextcloud targeted to my domain. I have forwarded ports 80 and 443 in my router to NGINX's ports (1880 and 18443) When I create and select a host in NGINX I am presented with an Error 522 page. According to the error Cloudflare say "The initial connection between Cloudflare's network and the origin web server timed out. As a result, the web page can not be displayed." Anyone have an idea where I go about addressing the problem? Edited August 13, 2021 by darrenyorston Quote Link to comment
plantsandbinary Posted August 14, 2021 Share Posted August 14, 2021 (edited) Bump, still having trouble with the above. Nothing is working. EDIT: Figured it out, my ISP died for 6 hours and I got a new IP address for the first time in about 12 months and this container doesn't work any more : Edited August 14, 2021 by plantsandbinary Quote Link to comment
Kopernikus Posted August 20, 2021 Share Posted August 20, 2021 Hi, I'm new to Unraid, but already have some experience with nginx on my Synology NAS. So I have a few questions about this docker container: - Whats the difference between this docker container and the one from "jc21"? - In your docker the network is set to bridge but in the "jc21" docker it's set to custom br0, what's the reason for this? - I have my own wildcard certificate so I won't be using Letsencrypt, so I have no plans of using http, only https, will the docker work if I only forwared port 443 to this docker container? Thx and keep up the good work 😀 Quote Link to comment
alturismo Posted August 20, 2021 Share Posted August 20, 2021 difference between bridge (only port on host is mapped) or custom br0 as sample (own ip with all ports available) is up to you, both should just work fine, you can change this how you like it, when you are on bridge you have to look for used ports from host or other dockers on that bridge, custom:br0 not, some have issues while using macvlan (from custom br0) and there is a new option in 6.10 therefore using custom with ipvlan, just in case ... 6 hours ago, Kopernikus said: so I won't be using Letsencrypt in terms you dont need http validation for certs using only 443 for https is fine of course. Quote Link to comment
jeremytodd1 Posted September 9, 2021 Share Posted September 9, 2021 (edited) Hey guys. I'm having trouble. I previously had NginxProxyManager up and running successfully, but I had to start over recently due to a mishap. I use Unraid, and I set it up exactly how I had it before., but I'm not having much luck. Currently, the only proxy host I have setup is for Overseerr (a tool to help request movies/tv shows). NPM says it's status is "Online" but when I try going to it it just gives me the generic "This site can't be reached". I can access it with the local IP address just fine. Here are a couple screenshots to help: Does anyone have any guesses? I'm not seeing a whole lot in the logs either. I have my password manager as well as NextCloud setup, so I (or my users) cannot access these services until NPM is back up and running successfully. Any assistance would be appreciated! EDIT: Adding this screenshot: Edited September 9, 2021 by jeremytodd1 Quote Link to comment
mattie112 Posted September 9, 2021 Share Posted September 9, 2021 Is your NPM container in the same 'proxynet'? If not use 192.168.1.216:5055 Quote Link to comment
jeremytodd1 Posted September 9, 2021 Share Posted September 9, 2021 5 minutes ago, mattie112 said: Is your NPM container in the same 'proxynet'? If not use 192.168.1.216:5055 Ah, I should have included that. Yes, NPM is also on proxynet. Quote Link to comment
mattie112 Posted September 9, 2021 Share Posted September 9, 2021 Hm... what error are you getting exactly? If you can reach NPM (but NPM not overseer) you should get a 502 - gateway timeout. Can you confirm you can reach your NPM? Does your domian resolve correctly, is the 80/443 port forward in your router correct? Quote Link to comment
jeremytodd1 Posted September 9, 2021 Share Posted September 9, 2021 58 minutes ago, mattie112 said: Hm... what error are you getting exactly? If you can reach NPM (but NPM not overseer) you should get a 502 - gateway timeout. Can you confirm you can reach your NPM? Does your domian resolve correctly, is the 80/443 port forward in your router correct? Yes, I can get to the NPM UI with no issue. Sorry if my post wasn't clear. NPM itself I can get to. The only proxy host that I have setup currently is for Overseerr. I can't get to overseerr.mydomain.com. When I try I get the "This site can't be reached" page in Chrome. 80/443 is forwarded to the respective ports that I have setup for NPM: Quote Link to comment
mattie112 Posted September 9, 2021 Share Posted September 9, 2021 If you get a "site cannot be reached" it probably not an issue with NPM but with your domain not resolving or your port not forwarded correctly. I assume you go to the NPM control panel through a local IP but your overseer domain resolves to an external IP. So just for sanity: Ping overseer.yourdomain Confirm it matches "whatismyip.com" Check your port forwards are they correct? If it all checks out you can change your hosts file to add something like "192.168.*.* overseer.domain.com" to force your PC to use a local ip. Just to verify NPM works that way. Oh and also check the NPM logs for any errors of course Quote Link to comment
jeremytodd1 Posted September 9, 2021 Share Posted September 9, 2021 (edited) 39 minutes ago, mattie112 said: If you get a "site cannot be reached" it probably not an issue with NPM but with your domain not resolving or your port not forwarded correctly. I assume you go to the NPM control panel through a local IP but your overseer domain resolves to an external IP. So just for sanity: Ping overseer.yourdomain Confirm it matches "whatismyip.com" Check your port forwards are they correct? If it all checks out you can change your hosts file to add something like "192.168.*.* overseer.domain.com" to force your PC to use a local ip. Just to verify NPM works that way. Oh and also check the NPM logs for any errors of course Pinging overseerr.mydomain.com does return the correct external IP address, per whatismyip.com The forwarded ports should be correct, as referencing these two screenshots: And checking NPM logs, I'm seeing literally no errors. For example, the logs in "NginxProxyManager\logs" are all 0KBs in size even, besides for error.log which just has this in it: 2021/09/09 16:01:17 [notice] 904#904: signal process started Edited September 9, 2021 by jeremytodd1 Quote Link to comment
mattie112 Posted September 10, 2021 Share Posted September 10, 2021 Allright, the reason why I am asking is if you can reach your NPM you shoud at least get somekind of error from NPM/nginx (like a gateway timeout). So that is why I am thinking that the problem lies before your NPM. (For example the port-forward in your router). You could try to run https://hub.docker.com/r/hashicorp/http-echo/ on port 80 or 443 to see if you get that output in your browser. Just to eliminate NPM from the testing to make sure where the problem lies. Quote Link to comment
jeremytodd1 Posted September 10, 2021 Share Posted September 10, 2021 I ended up figuring it out last night. NPM for some reason did not like ports 180 or 1443. After I switched those ports to 1880 and 18443 (and corrected the forwarded ports in my router) it started working correctly again. Quote Link to comment
spacezmonkey Posted September 12, 2021 Share Posted September 12, 2021 (edited) Hello After i install and check the log, i am having this error. is this normal ? When i setup host, i have error saving them. Edited September 12, 2021 by spacezmonkey Quote Link to comment
mattie112 Posted September 12, 2021 Share Posted September 12, 2021 No that is not normal, it seems to be in a restart loop. Did you make any custom config? Perhaps a IP list with no entries or something not sure. You can check the config file listed to see what is wrong exactly. Quote Link to comment
spacezmonkey Posted September 12, 2021 Share Posted September 12, 2021 Hello Mattie, Thank you for your reply. I do not make changes to setting during install as shown in my attachment. and i wasn't able to find the ip_ranges.conf file in " /etc/ngix/conf.d " . i do not even have the "include" folder after conf.d. Is there anything i have done wrong during the setup? Quote Link to comment
IngZ Posted September 14, 2021 Share Posted September 14, 2021 Lets Encript have folder /mnt/user/appdata/letsencrypt/www/.well-known Where this folder for NPM? Quote Link to comment
mattie112 Posted September 15, 2021 Share Posted September 15, 2021 20 hours ago, IngZ said: Lets Encript have folder /mnt/user/appdata/letsencrypt/www/.well-known Where this folder for NPM? Perhaps /data/letsencrypt-acme-challenge (in the container) ? Quote Link to comment
IngZ Posted September 20, 2021 Share Posted September 20, 2021 (edited) On 9/15/2021 at 7:35 PM, mattie112 said: Perhaps /data/letsencrypt-acme-challenge (in the container) ? not ;( any ideas? ps /data/letsencrypt-acme-challenge/.well-known/ Edited September 20, 2021 by IngZ ps Quote Link to comment
DieFalse Posted September 21, 2021 Share Posted September 21, 2021 I am having a problem with Argo Tunnel and NPM. Any client connecting to my environment, through https- gets logged as the Docker Network IP and not the CF Connecting IP. Has anyone gotten the Real IP to come through both Argo Tunnel into NPM? Its a config issue in NPM that I can not ascertain. Quote Link to comment
ShadowNovo Posted September 26, 2021 Share Posted September 26, 2021 Hey there, I've got a Nextcloud docker configured to be accessible at nextcloud.mydomain.com This works perfectly but when I just input "nextcloud.mydomain.com" into a browser it defaults to http, even though "https://nextcloud.mydomain.com" also works perfectly fine. How can I get Nginx to use https as default? Changing the "Scheme" to https results in the Docker not to be accessible anymore. Sorry if this is a rookie question but I'm trying to get this to work for an eternity now and did't find anything elsewhere using the GUI. Thanks and greetings ShadowNovo Quote Link to comment
Recommended Posts
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.