majorswitcher

the diskspace on the server I'm using for lemmyfly.org is getting full (it was 40GB local volume, and it's at 80% now) I can get extra volumes at my hosting (Hetzner) but is it possible lemmy looks at different volumes for the volumes folder that contains pictrs, postgres and lemmy-ui ?

Or do I need to move the complete main lemmy folder to this other volume and run lemmy from there, leaving my system-volume unused for lemmy ?

Pay for Sendgrid (100 / day for free is not enough) VS set up your own email server?

For own email server, do you not get flagged a lot as spam ?

I've set registration to Require application again, don't like this bots creating accounts.

Also I'm running into email limitations. Using the Sendgrid Free plan of 100 emails per day, but if a couple of member have email notification on, I'm reaching that 100 very quickly. How do others handle the high email traffic ?

running a lemmy instance has learned me a lot about devops already!

I love playing around with this, the lemmy instance was a great reason. lemmyfly.org doesn't have a lot of traffic yet, I think it can handle some more. Currently running on 2 vcpu/4gb ram.

After chatting to my devops colleague at work I wanted to set up a Grafana dashboard with metrics on the server. Seems it's pretty pretty doable!

• I installed prometheus node_exporter on my lemmy server
• I created a second server at my VPS. Very cheap, and don't think I'll be running the Prometheus/Grafana 24/7 so I'll just create a snapshot of it, destroy the server and re-create when wanted
• Created an internal private network so the instances can communicate without exposing ports to the public
• Installed Prometheus on the new prometheus server
• Installed Grafana
• log in to Grafana, admin:admin then change password
• Go to the Grafana home dashboard, click on + sign and click on Import. In Import via grafana.com, put the dashboard id 1860 (preset for all prometheus node_exporter metrics) and click on Load.
• Done!

Last 2 hours in the garden with a beer on the side, WFH FTW

cross-posted from: https://lemmyfly.org/post/2382

thanks to @Hopfgeist@feddit.de for noticing www.lemmyfly.org was not secured and not pointing to the lemmy instance.

Checking the lemmy ansible files I saw your nginx configuration file is located in /etc/nginx/sites-available/[sitename].conf

To add a redirect for your www.[sitename, add: (lemmyfly.org example)

server {
  server_name www.lemmyfly.org;
  return 301 $scheme://lemmyfly.org$request_uri;
}

check your config with sudo nginx -t

restart nginx: sudo systemctl restart nginx

Is it possible applications go missing? I received an email yesterday about a request and 2 hours ago, but can’t find them in the applications overview, neither unread nor all

When the admin of a instance wants to stop the server (for example costs are running too high) but your communities are active and you don’t want them to die along with it, is it possible to migrate the communities to another instance?

seems not scaleable. I thought you could subscribe at one instance, then browse communities from all instances and read and comment no matter if your instance has it in their Allowed list or not ? (apart from the ones in Blocked)

I've subscribed on my new instance lemmyfly.org to !lemmy_support@lemmy.ml. One post (by me) on there (postfix issues, possible to use sendgrid?) has 4 comments. But when I click on that post from my own server timeline, it shows an earlier state of the post with 0 comments.

Maybe I don't understand the federation well enough yet, but shouldn't the state of a post be similar no matter from which instance you look at it ?

edit: I see different comments on this post when looking at it direct on lemmy.ml/c/lemmy_support and on https://lemmyfly.org/c/lemmy_support@lemmy.ml

My own comment posted from lemmyfly.org, does not show under this post at lemmy.ml, only on my own instance ? This is why I edit the main post to write this. Is it supposed to work like this ?