My site certificate only supports non-www addresses. So, when we try to access an image uploaded to the site we are getting a bad certificate warning because there is a ‘www’ in the address. Is there a solution for this?

@nutomic@lemmy.ml
mod
link
fedilink
39M

How did you install Lemmy? And are you using pictrs for images? Because that is normally available through Lemmy (same domain).

@suspended@lemmy.ml
creator
link
fedilink
19M

I installed using these instructions: https://join-lemmy.org/docs/en/administration/install_docker.html

# address where pictrs is available
  pictrs_url: "http://pictrs:8080"
@nutomic@lemmy.ml
mod
link
fedilink
19M

Can you link your instance, and the url where the error happens?

@suspended@lemmy.ml
creator
link
fedilink
19M

Can you link your instance…

If you mean federate, then we aren’t comfortable doing that.

@nutomic@lemmy.ml
mod
link
fedilink
39M

No, thats not necessary.

@suspended@lemmy.ml
creator
link
fedilink
29M

OK. Thanks for taking the time to try and resolve this. I appreciate it.

@nutomic@lemmy.ml
mod
link
fedilink
19M

What value did you put in the line below? It should be beehaw.org, without www.

https://github.com/LemmyNet/lemmy-ansible/blob/main/templates/docker-compose.yml#L24

@suspended@lemmy.ml
creator
link
fedilink
19M

It is beehaw.org.

@nutomic@lemmy.ml
mod
link
fedilink
19M

Hmm, then did you configure www.beehaw.org in some other place? Because Lemmy certainly doesnt add that.

@suspended@lemmy.ml
creator
link
fedilink
19M

Here are the server blocks of my nginx conf file:

server {
	if ($host = beehaw.org) {
		return 301 https://beehaw.org$request_uri;
	} # managed by Certbot
	
	listen 80;
	listen [::]:80;
	server_name beehaw.org www.beehaw.org;
	location /.well-known/acme-challenge/ {
		root /var/www/certbot;
	}
	location / {
		return 301 https://$host$request_uri;
	}

}

server  {
	listen 443 ssl http2;
	listen [::]:443 ssl http2;
	server_name beehaw.org www.beehaw.org;
	ssl_certificate /etc/letsencrypt/live/beehaw.org-0001/fullchain.pem; #managed by Certbot
	ssl_certificate_key /etc/letsencrypt/live/beehaw.org-0001/privkey.pem; #managed by Certbot
@nutomic@lemmy.ml
mod
link
fedilink
19M

Its possible that certbot messed up something in your nginx config, but i dont see how that would affect the html sent by the server. Did you previously have www.beehaw.org set in docker-compose.yml? Maybe its still using the old setting.

Otherwise i cant think of anything.

@suspended@lemmy.ml
creator
link
fedilink
29M

Did you previously have www.beehaw.org set in docker-compose.yml? Maybe its still using the old setting.

I may have. I don’t know how it could retain the old setting. Strange. Thanks for helping.

@nutomic@lemmy.ml
mod
link
fedilink
29M

You need to run docker-compose up to reload the settings. With docker-compose restart, it keeps using the old settings.

@suspended@lemmy.ml
creator
link
fedilink
29M

I think I may have found something. Instead of using ‘restart’ with nginx shouldn’t I be using ‘reload’? See: https://linuxize.com/post/start-stop-restart-nginx/

reload: Gracefully restarts the Nginx service. On reload, the main Nginx process shuts down the child processes, loads the new configuration, and starts new child processes.

@nutomic@lemmy.ml
mod
link
fedilink
1
edit-2
9M

Either one should work. Anyway, this post shows that images are working correctly in new posts now. But you will have to edit old posts manually to remove the www from the url.

@suspended@lemmy.ml
creator
link
fedilink
29M

Thanks!

@suspended@lemmy.ml
creator
link
fedilink
19M

I’ve never used docker-compose restart. I always use docker-compose up.

Everything about Lemmy; bugs, gripes, praises, and advocacy.

For discussion about the lemmy.ml instance, go to !meta@lemmy.ml.

  • 0 users online
  • 1 user / day
  • 47 users / week
  • 40 users / month
  • 40 users / 6 months
  • 1 subscriber
  • 90 Posts
  • 480 Comments
  • Modlog