As far as I understand it currently people host their own Lemmy instances just for the hell of it or out of the goodness of their heart
But the larger instances will end up costing more money and I’m doubtful that will be sustainable with no income
Alot of the larger instances have user donations set up so that the servers can be maintained, this is how lemmy.world works
Crazy-ish idea, but maybe Lemmy could make a feature where instances can have custom awards similar to Reddit gold and stuff, and users can buy them to both award posts they like and financially support the instance host
I really like the overall concept of Lemmy, so I decided to set up lemm.ee to support the Lemmy network with my skillset. I have previously had the privilege of being responsible for running large platforms online (end-to-end, everything from operations to software engineering), and so far, this experience seems to be extremely relevant for running Lemmy in its current state.
As for paying for hosting, my initial plan was to to just pay for everything myself as kind of a hobby, but the userbase at lemm.ee has been very gracious in first asking me several times to share costs, and then actually sending money once I set up donations. I’m not sure yet if this donations-based funding will be sustainable, or if it will fall off after the initial hype dies, but for now it’s really awesome to see that there are several other people who believe in lemm.ee and want to share financial responsibility for it.
What does it actually cost? I have no basis for a ballpark guess even. I’ve seen this question asked to a number of admins and haven’t seen a direct answer.
It’s hard to judge how sustainable a donation based approach is without that info.
The costs will vary wildly depending on how the instance has been set up. If you set up all necessary services on a single VPS (as is the most common approach for smaller instances), then you can probably get by on $10-$20 a month. Splitting different services onto different servers, adding backups, load balancing, CDNs, redundancy, caches, etc will quickly increase the cost. Bigger instances need more powerful servers, that will increase the cost further.
On lemm.ee, we are currently not using very high-end servers, but we ARE using all the other things I mentioned above, and the monthly cost is currently hovering around $200 (that’s for 3 servers, a managed database, object storage, load balancing, a global CDN, and an e-mail provider). This is still on the very cheap side in the grand scheme of running online platforms, but definitely much more than I would want to pay for a single-user instance for example.
And honestly, 200 is on the high-end even with this setup. lemmy.dbzer0.com is way less
$200 is still on the low end, trust me - high end managed databases and compute resources are in the thousands, adding redundancy to that will double or triple it 😃
Ye if you’re adding like super redundancies etc sure. But I mean, it’s lemmy. We don’t need all that. My whole VM config is in ansible. I can literally scrap the whole thing, and redeploy it in 10 minutes. I just need to have a DB backup in case of some sort of catastrophic failure
IaC is awesome 👍 lemm.ee is deployed using Terraform.
My
lemmy_serverredundancy is mostly just so I can do infra changes without downtime (just take one node out of the load balancer, redeploy it and put the new one in), but it does also help a bit with general performance. I agree it’s not strictly necessary, the vast majority of other instances are just running on a single server, but I do think it’s very nice to have.For back-ups, I have point-in-time recovery, so I can restore the database to any random timestamp - and it has actually come in useful once already when two weeks ago I was able to restore to a good state about a few minutes after a problem - I think nobody even noticed that anything happened in that case 😃
Problem with Terraform is it assumes cloud providers, which tend to be expensive (or at least, I don’t know a way to do terraform on hetzner dedis :D )
My solution is to get cheapskate VPS and dedis and loadbalance them as frontends. The VM request is manual, but I only have to do this once anyway. It’s what I’m currently doing with the AI Horde. Of course, that doesn’t help when there’s DB changes but still.
What object storage are you using btw? I’m thinking to move to R2 or smt since I’ve had good experience with them until now. Contabo’s is way cheaper but when I tried to use it for high-demand stuff it dropped dead on the spot. But it might be ok for Lemmy.
I can’t thank you enough for sharing your knowledge, I am very interested in learning about server management and being able to read your thoughts is something I find extremely invaluable, please keep up with the great work! 😁
Yeah pretty much my setup. Full DB and pictrs backups uploaded to Backblaze B2 every few hours. Config files all backed up. Also send a copy to my home lab. Encrypted on the server before it gets sent out.
Worst case - we lose a few hours of data.
I keep about a week’s worth of backups on B2 (and 2 days on my home server) which is just over 200GB now. But that costs next to nothing on B2.
Maybe eventually if my instance grows more I might consider doing a replica DB on another server.
My total costs are like, $7.50 a month but I only have 15 actually active users. I don’t need to grow, but I’m willing to. If the costs increase too much I’ll ask for donations and if that stops covering things I’ll just close registrations. That’s the great part of not being a business chasing infinite growth.
I’ve been hosting a gaming server plus other related stuff myself for some years now.
While the user base will definitely be different, relying solely on donations is unfortunately not sustainable long-term. Donations fluctuate massively based on time of year in my experience. So it’s always good to periodically remind your community that lemm.ee needs donations to survive long-term.
When I do those reminders, users come out of the woods in droves to donate. It’s less that they’re unwilling to donate and more that they just forget to donate.
That’s great! It’s kind of a crowdfunded instance, then. Makes me wonder if it would be feasible to implement some sort of collection box plugin or something…
Yes … if anyone is a developer looking for ways to provide value to the fediverse … I suspect the donation process is probably of high value.
I don’t know the best way for it to be done … but something so that it’s easy for users to setup a single or regular donation and easy for devs and admins to put the relevant button right into their platform … all so that whoever is willing to donate has every opportunity to do so.
Yes, and I know it’s counter to the core motivations of this movement, but probably need a centralized repository for donation that can be a universal door for funds that can then be distributed to vulnerable, but active, instances. Needs to be run by a collective of reps from instances meeting a minimum threshold of support for the community. Also needs to be nimble enough to revoke funding is an instance takes a hard evil turn.
Or maybe just an app/site that recommends a distribution of a set monthly amount (e.g. 30 bucks) to the instances you use the most as a user?
I don’t think we need that at all. Each instance can look after its own hosting and funding.
A centralised pot is asking for trouble - what about theft or corruption? What happens about disagreements like defederation? Seems a lot more trouble than it’s worth.
I host my own. Specifically for myself and those who are friends or friends of friends.
I have a cluster of servers operating in my garage. Free real-estate for tons of stuff I want to host. I have to “pay” for electricity… the rest was already paid for long ago. My electricity cost for my whole cluster… is an estimated $1750 a year. But that cluster is 160 CPU cores, 750 GB of RAM, and ~400TB of storage. You ain’t getting that on a cloud hosted provider for $145 a month. About $110 of that is subsidized by my business operations. I host email, websites, nextcloud, plex, etc… boatloads of stuff.
You offer paid hosting with hardware in your garage? I’m getting dot-com flashbacks!
I actually do… Yes… I have dual internet connections, dedicated power off both phases of my electricity with ~5 hours of battery backup, run redundant internal infrastructure (power, network, and server hardware)… I also have a massive backup library and am currently working on obtaining offsite backups solution. The whole site/house is also monitored with cameras.
I have better uptime than some datacenters in my area have which I can truthfully quantify that as I also hold a CIO/CISO position in another company that operates out of a major Datacenter in my area.
Edit: Should clarify, an offsite backup solution that ISN’T “just peer with backblaze or some other provider” to store your backups. I intend to do a rotating tape library. With one set of tapes always being off site.
Also to mention… I have better uptime than AWS-east at this point of the year… Although there will be some outages of my infrastructure here soon for a network hardware update. I suspect something in the order of about 1 minute of downtime total.
obtaining offsite backups solution
I read this as
Put servers in my mum’s garage
That would be why I added my edit :)
deleted by creator
Mom’s spare bedroom closet 70 miles away, but yeah.
Gigachad is hitting five nines in their garage.
Sadly, Just lost it :( I knew the switch upgrades would do it… Eh. One more set of switch upgrades (40gbps trunk) here soon and they’ll be fully redundant too.
How does load balancing work across this fediverse… if at all?
What kind of load balancing are you thinking about/looking for?
If we look at Lemmy in specific…
Jobs that a server performs for users on it’s platform:
Login services(are you even you to begin with?)
Session Management (being logged in at multiple places, not mixing you up with other users…)
Subscriptions (what content to even show you… organizing them on the page… etc.)
Ban lists (and applying them to what you’re looking at)
Peering with the other platformsJobs that a server performs for off platform users: Sending ActivityPub messages to the other user’s platform.(effectively the same as “peering with other platforms” from above)
The idea is that the jobs that the server has to do for it’s local users is actually a lot more both numerically and in taxing tasks than the jobs that one particular instance has to do to send updates to the federated instances… A lot of the list in the first case is a boatload of SQL queries. The activitypub notification is effectively a broadcast of a relatively simple text that doesn’t take much bandwidth.
So let’s look at actual cases now that we have some ideas of what instances are doing.
I run a small instance, I have a few users (because I purposefully keep it small). If my users only ever interact with the fediverse through my instance, then other instances don’t have to do most of the jobs in the top list. I’ve taken that load from lemmy.ml or other bigger instances that my users would have ultimately migrated to if they didn’t have my server. The tradeoff is that my few users now cost the 1 connection for activitypub notifications.
Lets look at a theoretical “perfect” setup… 2,000,000 users across let’s say 1,000 instances. Evenly distributed. So each instance has 2,000 users. In that case, any given server would have to serve 2,000 people from the upper list and 999 activitypub broadcasts. This is significantly easier to do than have a single server try to handle all 2,000,000 users on a single instance serving nothing to the activitypub broadcasts.
There’s is one caveat with this… activitypub broadcasts will send everything that is subscribed… even if users don’t actually view it. So there could be some waste. This then leads to the discussion of well how many users until there’s a breakeven in cost of the first list vs the potential waste of the activitypub bandwidth. That answer is debatable… But most people agree that take the discussion seriously that the return on investment could be as low as 2-5 users. I think that the bandwidth cost outweighs the SQL (CPU, RAM, etc…) costs quite handily personally.
Other useful functions that occurs with this setup… As long as I got the ActivityPub message on my server… my users can see all of Lemmy.ml’s content. Regardless of if Lemmy.ml is actually running or not. We saw that with the downtime that some of the bigger servers are seeing, all the other instances could still consume the content… Just no updates were being sent.
This is relatively simplistic… I’ve skipped some nitty gritty stuff… I’ve simplified some other stuff. But this is the gist.
Edit: Other functions that you need to think about that lemmy is accomplishing for local instance users… Notifications, saved posts, language filtering, settings, profiles, avatars, reporting, slur filters, moderation, NSFW filters… The top list is actually considerably larger… like I said we simplified.
deleted by creator
I was wondering what Gilfoyle was up to these days
Why?
Seemed fun to do. Wanted to support something that gets people away from reddit.
How did I pay for it? I have a miniature datacenter in my house, complete with redundant power. Hosting lemmy is a drop in the bucket as far as my resources are concerned. As such, there really isn’t a measured cost for it. The infrastructure was already there and running, and lemmy doesn’t consume much of it.
I don’t take donations.
I’ve just launched one:
Why?
- I get free hosting with my job
- I’m a FOSS enthusiast and want to contribute to the fediverse
I hosted ad-free invision and phpbb forums for almost a decade before the centralized services started. I do it because I like social media and understand the value of a vibrant commons. I’ve been working with computers since I was a child so hosting yet-another-app-server is not really that big of a deal and gives my users and myself a place we can call home on the internet.
The out-of-pocket cost is minimal for a small instance, for larger instances scaling issues are going to hit eventually and costs will go beyond what 1-person will bear without complaint.
Highly recommend more smaller instances.
My server costs me only about $5 USD a month and I can host thousands of users without much additional effort (my sites before would usually run to ~10k users).
**Tl;dr we do this because we want to, the act itself is often fulfillment. **
I host my own because it’s cheap, less than 10 dollars per month, and I wanted to contribute to the growth of the Lemmy community.
Also turns out that it’s another benefit to this: I don’t have to get involved in the entire political debate of who federated with who. I just subscribe to communities I want anywhere.
I’m thinking about it. Do you need a domain name as well?
Yes you need a domain but that is usually quite cheap
Or use the raw IP :-D ?
You do need valid TLS and a cert can’t be directly issued on an IP.
Yep you need a domain name pointing to your instance when you run the installation of Lemmy and it will create a let’s encrypt https cert for your site automatically.
To have some control over my data and for the fun of it. With money? Selfhosting to me is a hobby and it has also taught me a lot, some of which has been helpful on my career.
I decided to host my own because I was on lemmy.world and we got blocked by beehaw, which had many of the communities I wanted to be a part of. I run mine on a server that’s in my house, so the only thing I’m paying for is electricity. And I have solar panels.
How difficult is it to host your own instance? Can you still use apps like Connect with it?
Yes to the last question
It was really annoying to set up if I’m being honest. If I hadn’t taken classes on Docker, I would have never figured it out. Luckily they have been improving the process recently. It already much easier now than it was a week ago. Hopefully by the next major release it is easy peasy.
Clever
lol I was the same way. I went from not being able to see just beehaw to spotty federation with many other instances (oddly enough beehaw and the kbin instances are nearly perfect), so I’m not yet sure if it was an upgrade.
I host my own instance for the same reason I self-host the dozens of other services. To have control over my digital services the best I can. I have a few server machines running various services. I run like 40 docker containers running at the moment. Lemmy is a set of those containers.
It cost me the electricity cost to run the server, plus the cost of my internet. I suppose you should include the initial hardware cost- my servers are basically my old gaming rigs. Not to mention the time investment to maintain another service. For me, it’s worth it to self-host if I can.
Specifically for lemmy, I seen how overloaded the various major lemmy instances out there were, so self hosting could mean one less user on those instances. I also didn’t see any significant drawbacks to self-hosting the instance since I can still join and communicate with all the other communities.
I can see that, but I expect the main cost of servers are the hosting of subs, not maintaining users.
I expect that those servers are going to need support eventually.
I expect the main cost of servers are the hosting of subs, not maintaining users.
Which is easy to avoid, if you want. You can disallow the creation of communities on your instance.
So users can still join and make it their home, but they will have to subscribe to communities on other instances.
I’m not sure about the ethical implications. It means you’re sort of leeching off moderation and maintenance from other admins.
I guess it’s a negligible point and totally fine, but not sure. Would be interesting if community-admins could share their thoughts on this.
But if everyone avoids hosting communities, then Lemmy is going to die because no one is going to be in a community.
It may not be an ethical issue, but it is a financial one that needs to be addressed.
I tried out my own, one person instance, and I had a hard time getting comments to pull in. I would have to browse to the original instance to view all of the comments. Have you found a good way to overcome that?
I switched back to a more public instance just because I found the process of going out to view the content, back to my own instance if I wanted to comment, then back to the original again to keep reading the discussion very tedious.
Being on a more populated server seems to give most if not all of the comments directly.
When you first subscribe to a community it only pulls in the last 20 or so posts and I think a limited amount of comments, and then everything going forward. This seems to be a common point of confusion for a lot of instance admins.
Presumably this is to prevent a possible DDoS/performance failure vector as it would be trivial to setup a large swarm of instances on tiny VM’s and then simultaneously start hitting massive communities from a single instance and requesting a large body of historical content.
Edit: Also when you first setup and start subscribing to a large number of instances, this initiates a LOT of communication and database writes. Lemmy still has some performance bottlenecks. Once everything is initially synced and settled it runs fine. I have a friend running their instance on a $5 Linode instance that only has 1vCPU and 1GB of RAM without any issues, and they’re hosting users.
I ran mine for a couple weeks, and communities I’d been subscribed to from day 1 were still missing most comments on the posts unless I clicked through to their page. Maybe there was something funky with my install, but I used Lemmy’s ansible scripts to deploy so I don’t know what else I could do.
I’ve loved the idea behind Lemmy since I first discovered. At first, I was using lemmy.ml, but then I saw the opportunity to provide a nice space and expand my sysadmin skills. Since there was no Portuguese instance yet, I thought why not create one?
Since then, I’ve met more people hosting Portuguese services and it has been great :DFor funding, I’m working on two ways: the typical donations and trying to secure support from local FOSS organizations. At the moment, the server costs are not prohibitive and there have been some donations already. I’ve also been talking to some of those orgs and it’s going well :)
This sounds encouraging, I hope your instance flourishes
Thank you! :3
I don’t see why an instance couldn’t go with ads and subs or awards the same way Reddit does.
Because they’d probably have to create that tech themselves. Not that they won’t but it’s not there ATM afaik
I have a 4 core 24 GB ARM VPS on that Oracle free tier that I use to self host some personal services.
I just think it’s neat.
What’s the catch? How long does the free tier last?
There’s always a catch. Oracle has and will claw back that free allowance unless you meet specific criteria or move to a pay-as-you-go tier with a cc on file with a $100 authorization charge, and even then they may. Ask me how I know lol
Or they’ll just shut it down with no explanation whatsoever, even if you’ve met all of their weird criteria.
There are two classic blunders. The most famous is “never get involved in a land war in Asia”, but only slightly less well-known is this: “Never go into business with Oracle when anything of value is on the line”.
Never go into business with Oracle when anything of value is on the line
Truer words have not been spoken.
That’s not even considering that you’re not going to be able to talk to a rep about it, or appeal that claw back in any way. Hell they’ll even terminate your whole account without warning, not just the VM. If I found out the admin of the Lemmy instance I’m using was hosting it on Oracle I would delete my profile immediately without a second thought.
If I found out the admin of the Lemmy instance I’m using was hosting it on Oracle I would delete my profile immediately without a second thought.
Agreed. I still feel dirty for taking Google’s side in that ridiculous lawsuit Oracle filled after buying Sun. Only Oracle would try to make money from licensing fees for APIs…wait a minute…has Spez been talking to Elon Musk AND Larry Ellison?
One other catch is that is an ARM instance
Oh shit really? You get that much processing power for free? I gotta sign up to that
Except that the instances are usually full. There is hope, but not much
Me too, works well!
is this a dedicated vm or something else?
In the medium term I’m confident costs could be met using patreon or something similar for large instances.
Plenty of people willingly paid for reddit awards etc. I think most of us feel more loyalty to Lemmy than reddit in the light of recent events.
I am already hosting the AI Horde so I’m good with infrastructure. And since lemmy provided an ansible playbook which is my expertise, I thought it would be easy. And it was, even with my using my own loadbalancer solution instead of nginx.
I already had a server I wasn’t using much for the AI Horde so I repurposed it. The DB physical I have is really good so I just made it share the DB for lemmy as well.
I pay for it from the same budget which runs on donations. I put a Kofi link on the sidebar and already covered so much of the costs that I sent some of the overflow to lemmy development
I sent some of the overflow to lemmy development
I was hoping for something like that the other day. Makes me happy to see it happening! Thank you :)
Word to the wise, Lemmy devs prefer liberapay over patreon and opencollective.
Do they explain why?
Iirc, liberapay take a 0% cut.
I like that open collective has public auditing though, which I think is important for FOSS projects.
You may well be correct, I simply read a comment that said they prefer Liberapay.
like many others, I hosted my own for a variety of reasons.
- I like to tinker
- this was a chance to do a lot of tinkering and get in on the “ground-floor” of something I think has a lot of potential
- it gives me a bit more control over the experience, which is helpful when trying to get friends/family to give Lemmy a chance.
As for the cost, I have several VPS’s which cost ~$40/year, so the cost is basically negligible.
$3.33/month? I too would like to know who your VPS provider is.
Can I ask who your VPS provider is?
