Fly.io seems to be a bit of a mixed bag:
https://news.ycombinator.com/item?id=41917436
https://news.ycombinator.com/item?id=35044516
https://news.ycombinator.com/item?id=34742946
https://news.ycombinator.com/item?id=34229751
If a cloud platform doesn't really provide reliability, I'd say it's probably not worth it. You could better just rent a (virtual) server and save the cloud tax.
For experiments and hobby projects the value proposition is amazing. Where else can you spin up an independent instance for $1.94 per month?*
*Note this is for an instance with only 256MB RAM (https://fly.io/docs/about/pricing/), but it's definitely possible to run non-trivial projects on that. Rust-based web servers like Rocket require only about 10MB RAM. Basic PHP servers should also fit from what I can find.
There are plenty of better deals as long as you don’t limit yourself to big clouds and clouds with startup-esque landing pages frequently posted to HN. LowEndTalk may be the most well-known place for finding such deals.
(Not saying the typical cheap VPS on LowEndTalk has comparable PaaS features. Only responding to parent’s use case of a single cheap instance.)
Best business model in the world, buy stuff in big bags, put it in smaller ones, sell at a multiple of the original price.
Fly is mostly (to my knowledge) reselling Netactuate and OVH servers, their main innovation is the developer experience on top, using Docker on a MicroVM based approach. Of course not only that, but I think it’s their main differentiator.
Haven’t used that in a while but Scaleway offered ridiculously cheap dedicated ARM hardware close to these price points, not sure if they still do.
Nowhere? Because that's a ridiculously low amount of RAM to offer even in your cheapest offerings?
You can easily get 4 GB of RAM for $5 from the likes of Hetzner or Hostinger, so that's 16x more RAM for 2.5x the price. One relatively unknown provider I have used in the past offers 2 GB of RAM for €3.6/month (if paid monthly, €3 if anually), so 8x more RAM for 1.5-2x the price. I'm sure I could find something even cheaper, but I'm just looking at providers I have personally used.
BTW that dropdown seems to be sorted cheapest > most expensive. If you go to the bottom of the list the price for that same VPS doubles.
> Nowhere? Because that's a ridiculously low amount of RAM to offer even in your cheapest offerings?
There's definitely places that offer it... also 512m
I know because I've personally bought such plans and that was $5-10/yr because I didn't need dedicated ipv4.
Oracle free is one 4 core 24gb ram vps + 2 dualcore amd vps.
And actually, it's the resources that are free (CPU, memory, network) and you're allowed to split them up into multiple VMs if you want to.
One of my VMs had an uptime of more than 1050 days before the infrastructure rebooted it, so in terms of availability they've certainly surprised me.
The only downside I've come across with Oracle Free is that the 'best' regions are typically full. I ended up provisioning my free VMs in another region/country and it works fine.
I suppose another downside (if you want to view it this way) is they will delete idle unused free VMs after a certain time period. You have to add a credit card to your account to "upgrade" your account and run free resource indefinitely. While you're not charged for anything, it makes me nervous forking over a CC number to Oracle.
One such microVM per month used to be within the free monthly allowance, is that not the case anymore?
Maybe if you're limiting yourself to AWS-wrapper cloud companies. What good is a $2/mo cloud instance if it's down multiple times a month?
Just get a $5/mo VPS instead if you're really concerned about a few dollars a month.
> What good is a $2/mo cloud instance if it's down multiple times a month?
The perverse irony is that the most common reason cited by cloud providers for not letting people set a hard cap on charges is an insistence that surely the last thing you want in the world is for your service to be taken offline, even if it does means avoiding a $1k–$100k bill at the end of the month.
I used to use Racknerd for that sort of thing, and the costs were around there -- maybe $1.90/mo for a 512MB instance. It was easy to squeeze several hobby projects onto the machine.
i recommend lowendtalk what fly.io doing is running colocated baremetal servers and using firecracker to overcommit (probably via memory ballooning and other disk compression on demand)
if you are going to haggle over $2/month then you are better off just connecting your raspberry pi with wireguard/cloudflare tunnel on a residential connection
The reliability is very very bad. It was really insane that 2 times in the past few months the main dashboard was down as I’m demoing something. Not to mention the deploy outages and almost daily some random thing was unavailable or delayed.
I had to leave a few months ago after the price raises and how many times my boss saw some issue in the project I had with them.
They also deprecated and removed their sqlite backup service. Back to GCP and not worrying about so many outages now.
Now just to worry about GCP getting shut down with a few days' notice. /s
But in all seriousness the gall to raise prices before actually fixing the reliability problems is pretty shocking. I understand it's a bit of a chicken-and-egg thing where you maybe are tight on resources but there's no scenario where it's acceptable to have a product with these kinds of problems and then raise prices on existing customers who are putting up with it.
theres just so many anecdotes/nightmare stories from people using fly.io here much more than the ones linked by GP
expect to see more of these "post-mortem apologies" from fly.io in the future because it won't be the last
I don't really understand the value prop of fly.io. They seem to have an impressive engineering team despite the outages, but is edge compute really something that 99.9% of devs need? There are tons of large companies that operate out of a single AWS region and those services are used by millions around the globe. It just strikes me as something that enables premature optimization right out of the box.
It's basically the new Heroku with less lock-in, because it works with Docker.
You get edge computing, autoscaling, and load balancing without additional configuration.
Not as flexible as AWS, but also much easier to setup and maintain.
But the reliability issues suck now and then.
> Not as flexible as AWS
Today, Fly.io is more or less in the same market as Lightsail, not AWS. And when you compare it to Lightsail, it blows it away.
Did you count reliability into your assesment here? I'm reading about Fly.io outages multiple times a year, whereas Lightsail seem to be as stable as AWS EC2.
And when you compare it to Lightsail, it blows it away.
This is a bit of a confusing sentence because there are so many pronouns. Do all of the "it"s refer to Fly.io?
> And when you compare [fly.io] to Lightsail, [fly.io] blows [Lightsail] away.
DigitalOcean has been doing this for years, and their value proposition is unmatched IMO
For $5 you get:
Latest gen CPUs and RAM
HTTPS
DDoS protection
Cloudflare CDN
Autoscale
Competent support
I'd say the best part is the predictable monthly prices
And while most people probably don't care, they are an established public company, so there is more chance they will exist in 10 years
are global r/w token permissions still a thing, or did the token scopes thing finally come out of beta?
also, my experience with support was not the same as yours. they were utterly useless for the most part.
for a personal web dev (or similar) project, like, i agree, they’ve got good value.
but having worked in a small biz where DO was what they built everything on — no. bad idea. spend more. use aws (graviton ec2 instances)/azure.
the $5 droplet is underpowered and can't run anything substantial. it's just the price to get you in the door.
It doesn't really need to run anything "substantial" though. Running some janky wordpress site with some scabbed-on ecommerce customizations is like 50% of the internet.
a 1vCPU 512mb instance is plenty for most base cases. Maybe you need one additional machine to act as a background worker. I am sure there are some noisy neighbors but to say its underpowered is silly.
I'm calling it underpowered because the $5 one had trouble running my custom ssh daemon. ssh! the cryptography for that shouldn't chug down the server I'm renting from them. a bigger instance from them isn't having the same problems.
you wouldn't be able to run anything substantial with that kind of budget
but GO and pocketbase is on record for supporting 10k concurrent requests per second on low powered VPS
This is precisely it. The ease of deploy, https domain configuration, scaling.
Additionally, having machines that turn off when not in use is easy to configure, which I never managed on AWS.
> which I never managed on AWS
I haven't looked at it recently, but App Runner could do a few of Fly.io esque things (but slightly more expensive): https://aws.amazon.com/apprunner/
I have asked this multiple times but is anyone really using edge compute and getting value out of it? I am certain there are cases but I have not seen any of them written up before.
Depends on what you mean by edge compute, but you probably are.
5G towers are a ton of compute on the edge to secure and protect the traffic passing through them.
Or if by edge you mean having stuff close to your consumers, every non trivial operation does that.
How is it not obvious based on the thread at hand, fly.io.
And no not every nontrivial operation does it to the extreme of an envisioned fly.io deployment.
We have an embeddable audio player served globally with very low latency. This wouldn't be possible without edge compute/data.
I have an SSR Astro project. Using Fly makes my project fast.
For dynamic data I use SWR.
I could use Cloudflare workers but it doesn’t play so nice with Astro.
I also have a “form submission service” where I receive a Post and send an email.
I need maximum uptime to avoid revenue loss.
It’s a go service so I deploy ~6 machines across the US to ensure I don’t drop any requests.
I haven’t had downtime in years.
If half your customers are in new your and half in sidney it makes you app faster if you run it in both places.
There is a lot of things we do for our users that we don't need (no one "needs" SPA etc). But if it is easy to make your app faster for your users, why not?
I would take edge compute if it's free and easy. That's fly.io's value prop.
In a world where much web browsing starts with ACK SYN ACK, it is nice if the server is close to you.
I typed fly launch, fly deploy and my node.js project was deployed. So I guess hobby projects?
I am going to go out on a limb and say there is no real value prop to fly.io. I could completely be wrong but it always feels like the modern MongoDB. Everyone wants to use it but I am not sure they are extracting value from it and instead its a shiny toy that is fun to build from.
This is a completely sane way to look at the world and we won't push back on it at all. We're building something extraordinarily difficult, and we're a relatively new company, and we don't have even a fraction of the resources the hyperscalers do, or, in the cases of AWS, GCP, and OCI, had at the time they started. If you're minmaxing for reliability --- which is an absolutely sane way to play --- we're not going to tell you you'd do worse in 2024 UE1.
If it helps: all sorts of things can and do go wrong, but the most likely form of disruption you're likely to see here are periods of times when deployments don't work. This outage was a deployments/orchestration outage. We had a total request routing outage several months back, owing to a Rust concurrency landmine we stepped on, but those are very rare.
(Deployment and state-update outages are a big deal, and if you deploy to diverse groups of Fly Machines constantly, as we encourage you to do, that being one of the big features of the platform, they can impact your availability. I'm not downplaying them.)
fly.io has a very bad reputation for reliability there doesn't seem to be any damage control beyond hackernews and even here the consensus seems to be "dont run anything mission critical on fly.io or expect data redundancy"
in fact, you can almost get the same thing fly.io does by running firecracker on your own bare metal servers and cheaper too.
I'm afraid the public sentiment towards fly.io has been tainted for good (I can't count how many times they apologized now).
This is the second place you've offered this sentiment. Was it your expectation that we were going to hit some point, sometime in the near future, where we weren't going to have deployment-blocking outages? I'd like to better understand your premise. If it's "I can get more reliability by deploying on a hyperscaler cloud", who ever told you otherwise?
I see so you think its good business practice to basically say "expect more downtimes in the future who cares about your entire business going down for several hours more than once a year.
Gotcha. I'll be sure to pass on the good word.
You'd be happier with a comment saying "there will be no future outages", I see.
I tried out Fly.io and deployed a little test app. I couldn't even access the app, because they put it onto a server that was under "emergency maintenance" and had been that way for twelve days.