lemmy.ml meta

1406 readers

1 users here now

Anything about the lemmy.ml instance and its moderation.

For discussion about the Lemmy software project, go to !lemmy@lemmy.ml.

founded 3 years ago

MODERATORS

nutomic@lemmy.ml

172

lemmy.ml is migrated to a new server (lemmy.ml)

submitted 1 year ago by nutomic@lemmy.ml to c/meta@lemmy.ml

102 comments fedilink hide all child comments

Its now running on a dedicated server with 6 cores/12 threads and 32 gb ram. I hope this will be enough for the near future. Nevertheless, new users should still prefer to signup on other instances.

This server is financed from donations to the Lemmy project. If you want to support it, please consider donating.

you are viewing a single comment's thread
view the rest of the comments

[–] sealneaward@lemmy.ml 0 points 1 year ago (1 children)

Is most of the discussion about the infrastructure done in a community here, or do you folks have an invite only discord/slack or something @nutomic@lemmy.ml?

I've been working in devops for 4 years now, but with mostly kubernetes and AWS, but I'd like to throw some ideas and just ask some more specific questions about the infrastructure. Like I'm curious why autoscaling the web servers and moving the database server to a dedicated instance is not the current configuration.

[–] nutomic@lemmy.ml 3 points 1 year ago (2 children)

There is really not much to discuss. The server was overloaded so we got a bigger one. And there is no reason to mess with stuff like kubernetes when a single server works fine. After all our job is to improve the Lemmy software for everyone, not build a huge centralized platform only on lemmy.ml.

[–] qprimed@lemmy.ml 1 points 1 year ago* (last edited 1 year ago)

Exactly. lemmy.ml is lemmy.ml - just a single instance. there should be incentive for other to stand up instances of their own. overload on any instance is good reason for ppl to get out there and build new instances.

[–] sealneaward@lemmy.ml 0 points 1 year ago* (last edited 1 year ago) (1 children)

I agree with the kubernetes point. Not every problem needs it.

I was more curious about having a load balancer infront of an autoscaling group of servers with the existing images that you are running off of, minus the self hosted database server. Then you would be able to handle spikes automatically. Just curious if that has been thought of.

Having it accessible to everyone is great. For sure. But I was just thinking that having stronger pieces of infrastructure for instances that handle more traffic might be beneficial.

[–] nutomic@lemmy.ml 0 points 1 year ago (1 children)

So far it's not necessary and there are more urgent problems to fix.

[–] sealneaward@lemmy.ml 1 points 1 year ago

I agree, lots of application improvements and bugs are probably your priority. That should always be the focus of the application developers. I honestly want to try and experiment with this idea now.

If I get a good repository going that focuses just on the infrastructure side of things, I will post it here.