aptalca

"as" stands for "access server", and is the gui version by openvpn. You need an openvpn client to connect to an access server. Linuxserver.io doesn't provide a client image. No idea about the other docker images

Nginx is not listening on port 80 by default because mapping port 80 is optional for this image (unless you're doing http validation) The readme states that listening on 80 for http and redirecting to https can be enabled in the default site config

Define not working. Post your config. It works fine here. Read the last few pages here for more info.

No idea about unms built-in letsencrypt so can't help you with https://unms.mybusinessdomain.co.uk. When you use the 6443 port, you are bypassing the letsencrypt container and connecting directly to unifi, so no proper cert, only self signed, hence the warning message. With regards to unifi.blah address, use the proxy conf for it, make sure you only have one active unifi conf, set the server name to that full unifi domain url. Also noticed that you changed your container name to "Unifi". That won't work as nginx won't resolve names with uppercase letters. Make sure your unifi container is named "unifi". Just keep the unifi proxy conf default except for the server name directive.

Fyi, the container now contains its own interface called eth0 inside. In bridge networking, this eth0 network gets bridged to the host's interface.

Exposing guis like sab and radarr is the exact reason for a reverse proxy. As long as a proper auth is used (http auth with fail2ban as implemented in this image) it's perfectly fine.

If you change the set, you'll be able to get a cert because your current restriction is "for the exact set of domains". I recommend starting with one duckdns subdomain as the url and see what happens. You can then expand if you need to

Unfortunately the serviio image has been deprecated. The existing image will remain on docker hub but there will be no further updates or refreshes

It seems when you messed with the permissions, you prevented the container from properly deleting the certs folder and when it regenerated certs, it created a duplicate folder with a 0001 appended to the name. Start over, don't change any permissions. Also a side note, duckdns shouldn't really be the url variable. Url should be yoursubdomain.duckdns.org and subdomains should be any sub-subdomains you would like to use ie. "sonarr" so you can access sonarr.yoursubdomain.duckdns.org If you need to add additional duckdns subdomains (usually needed for servers with different IPs, although I doubt that's the case here since they all need to point to the same letsencrypt instance), put them in the extra domains.

It's the last line in the config file you posted, the line that starts with proxy pass

Post your container settings

Just curious, what are you guys plus one'ing? Docker compose is already added to the nerd pack plugin. It's a one click install

Then like I said, use port 80 in your proxy pass directive

Is your speedtest running inside a container named speedtest? And is that service listening on port 6580 inside the container? If it's actually listening on 80 inside the container but you mapped it to 6580 on the host, you need to use 80 in the proxy pass directive

It's described in the readme: https://github.com/linuxserver/docker-letsencrypt/blob/master/README.md#validation-and-initial-setup

The image doesn't set anything to read only. Provide more info on what you're experiencing

Your screenshot shows host networking with bond0 as interface

Did you map the ports?

When you transcode hdr, the colors will be washed out, no way around it. Google plex hdr tone mapping and you'll see plenty of articles and posts about it. With regards to buffering, you must have a bottleneck somewhere, you are working with a high bitrate file, so it could be io or network related as well

There are samples in the default site config that are barebones. Set and resolver lines are used in conjunction with connecting to other containers by container name. They are not needed when ip address is used, but wouldn't hurt if left in.

?? With compose, you just do dc pull and dc up -d How do you update and clarify how it used to work but it no longer does?

Yes, use the server ip and port on the proxy pass directive

Excellent, thanks

Can we get an update for docker-compose? Current version is 1.23.2 Thanks

I an not trying to be mean. You said you couldn't find the info anywhere so I pointed you to it. Just making sure you have the info. We only update one place, GitHub (readme), which is the ultimate source of both the code and the documentation. That automatically gets synced to the docker hub with each image push, and is the ultimate source of our images. Everything else links to those two places because we don't have the time to update so many different sources every time something changes.