mattie112

Is there not an environment variable which has the path to whatever folder it consumes from?

But if you use ipv6 you kinda do need to use it

I made a fork for that: https://github.com/Mattie112/docker-nginx-proxy-manager (I update it a few times a year if I dont forget haha)

Open the console to NPM via the Unraid container page (next to stop/start/restart) and remove it using the shell. OR: delete NPM, then re-'install' it and use the same config/path mapping.

Manually remove the log try again if you already increased the size of the docker image.

The files are not stored in the container but mounted from the /appdata storage share. If you want to start over / clear everything empty-out that folder. Check your docker config to see the exact path of mapped volumes.

And other then what was already suggested: try testing things one by one. Can you access your NPM over IPv6? Does the DNS resolve correctly (perhaps your Unraid machine has a different IP then your NPM container) Does the forwarding work over IPv4?

For quite some time now a docker pull is painfully slow on my Unraid machine. I looked at and already but that seems to be good for me if I understand it correctly. Any ideas on what I can test? tower-diagnostics-20241026-1958.zip

You could try two things. With your phone on 4G try to access both 80 and 443. Are both not working or just one of them? Just to make sure we are looking in the right direction.

Ik strange. You could check in the docker container itself on the filesystem (or in the app data) and manually remove any files that are certificates for your host. Or if you have not a lot of config yet just clear and restart.

Can you chech the certificate view in NPM? Perhaps it did made one and you can just use it. Or chech the drop down where it says 'request new'

You can technically migrate certificates. But: you don't need to, if everything is set up you just get new certificates from letsencrypt. What exactly is the problem? And how is it set up?

Ah, that wasn't so bad Everything has been updated again if someone wants to use it.

I do have a fork that only changes the ports to 80/443 that I have been using for quite some time. (And I think also a couple of other people). Let me update it real quick with all the latest changes as I have not done that in quite some time https://hub.docker.com/repository/docker/mattie112/docker-nginx-proxy-manager I will report here when I updated / tested it, will be somewhere today.

If you mount a file in docker it will always overwrite what is present. (but feel free to create a PR). I did do this way, I think if you search in this topic you should find my post with the file I use.

I am currently on holiday and I can't really go through a zip on mobile . I want to suggest to open the docker logs of NPM and see if it logs anything when you open your site. Can you check and paste a bit of log here if that's the case?

You can create your own config file and "overwrite" that with a docker volume. I do the same and it works perfectly.

Or run it on a different IP all together. That should also work (so bridge network)

The logs?

You cannot put a file in a SMB share? Or your paperless has problems accessing a folder? The first one is not a paperless issue so perhaps worth a separate post? In the Unraid Tools menu there is a "(docker safe) new permissions" option that might fix your issue.

Try to renew it via de CLI (see elsewhere in this topic posted it numerous times). There is a --dry-run or a --force-renew flag for certbot. Then see what errors you are getting.

No problem! That is the main reason I use it. Services don't need to bother with SSL/certificate. It is all handled by NPM.

Internally you can use Http just fine! Perhaps only if your NPM and service are on physically different machines then it is a question of: do you trust your internal network.

So: This is still an issue for me. Had a very long uptime, recently restarted due to an upgrade -> no problem. Now I had a "crash" (again due to filling-up the cache dir). Could not recover, removed the docker image and had it recreate. Then re-added all my containers and behold the same problem again It is exactly the same as in the first screenshot, however disabeling and then re-enableing docker does not fix the issue right now. docker run -d --name='NginxProxyManager' --net='br0' --ip='192.168.100.201' --ip6='xx:xx:xx:xx::xx' --privileged=true -e TZ="Europe/Berlin" -e HOST_OS="Unraid" -e HOST_HOSTNAME="Tower" -e HOST_CONTAINERNAME="NginxProxyManager" -e 'TCP_PORT_8181'='8181' -e 'TCP_PORT_8080'='80' -e 'TCP_PORT_4443'='443' -e 'USER_ID'='99' -e 'GROUP_ID'='100' -e 'UMASK'='000' -e 'APP_NICENESS'='' -e 'DISABLE_IPV6'='0' -l net.unraid.docker.managed=dockerman -l net.unraid.docker.webui='http://192.168.100.201:8181' -l net.unraid.docker.icon='https://raw.githubusercontent.com/jlesage/docker-templates/master/jlesage/images/nginx-proxy-manager-icon.png' -v '/mnt/user/appdata/NginxProxyManager/matthijs-custom-ssl-ciphers.conf':'/etc/nginx/conf.d/include/ssl-ciphers.conf':'rw' -v '/mnt/user/appdata/NginxProxyManager':'/config':'rw' 'mattie112/docker-nginx-proxy-manager' And then the reply is: ``` Error response from daemon: user specified IP address is supported only when connecting to networks with user configured subnets Error: failed to start containers: NginxProxyManager ``` I tried removing br0 and manually creating it: `docker network create --driver=macvlan --subnet=192.168.100.0/24 --subnet=xx:xx:xx:xx::0/64 br0` That allows me to start the container but no network traffic possible. So yes: any help is appreciated again I could also open a new topic but this is exactly the same problem so yeah.... (or at least to me it looks like the same problem) edit: Tried changing to eth0 as listed in changelog (https://docs.unraid.net/unraid-os/release-notes/6.12.4/) no luck!

Those header are really application headers. Yes your web server can send out headers but is more for your application to define them. Your webserver does not know what your application wants. But yes: you can "override" them in nginx.