Skip Navigation

[Done] Server will be migrated (More power!)

So after we've extended the virtual cloud server twice, we're at the max for the current configuration. And with this crazy growth (almost 12k users!!) even now the server is more and more reaching capacity.

Therefore I decided to order a dedicated server. Same one as used for mastodon.world.

So the bad news... we will need some downtime. Hopefully, not too much. I will prepare the new server, copy (rsync) stuff over, stop Lemmy, do last rsync and change the DNS. If all goes well it would take maybe 10 minutes downtime, 30 at most. (With mastodon.world it took 20 minutes, mainly because of a typo :-) )

For those who would like to donate, to cover server costs, you can do so at our OpenCollective or Patreon

Thanks!

Update The server was migrated. It took around 4 minutes downtime. For those who asked, it now uses a dedicated server with a AMD EPYC 7502P 32 Cores "Rome" CPU and 128GB RAM. Should be enough for now.

I will be tuning the database a bit, so that should give some extra seconds of downtime, but just refresh and it's back. After that I'll investigate further to the cause of the slow posting. Thanks @veroxii@lemmy.world for assisting with that.

354 comments
  • I really appreciate what you're doing, but I'm worried how this instance will continue scaling. What happens when it gets to 1 million users? 10 million? We can scale vertically only somewhat, but horizontal scaling seems to be limited to "just join a new instance 4head" and that just...doesn't have a good experience.

  • I'm not an engineer or a dev - but requiring a 32-core, $2000+ CPU to support 12k users doesn't seem like it would scale well. Is this normal, or does the fediverse require more computational resources than a simpler setup like reddit? How would a fediverse instance with 100k users be maintained?

  • do you plan to publish any of your scaling data? Some others might consider helping by running large instances and your learnings would be incredibly helpful.

    • Yes, when I get around to it I can create a post about it. And of course, feel free to ask.

      • if you had any basic numbers easily accessible like iops peak and average or screen shots of your dashboard showing resource utilization would do wonders. Right now you are getting slammed but are operational (barely) you have the upper bound config for a 12k user system at least right now ;-)

  • Thank you very much. The welcome for all us reddit refugees has been really warm and it's deeply appreciated.

  • Thank you for making this happen! Just signed up for a regular donation to help with costs 👍

  • I’m just another reddit refugee but I wanted to say thank you for your time, effort and money. As I slowly come to terms with federation I see why some are so passionate about it.

  • Sounds like Lemmy is quite power hungry. What are the specs of the new server? Can Lemmy be split into its services (e.g pictures on a different servers?)

    • Pictures is already it's own service in the docker compose file. I don't know how easy it would be to move it to a different server though, but I suspect it can be done with a bit of fidling with the configuration.

      • Also, it should be fairly easy to host the postgres on separate hardware.

  • I don't know what most of that means but you guys (and/or gals!) fuckin rule and I appreciate the absolute hell out of you.

354 comments