mrkurt

25 ms·

271.
by mrkurt 4y ago
The downside of v8 isolates is: you have to reinvent a whole bunch of stuff to get good isolation (both security and of resources). Here's an example. Under no circumstances should CloudFlare or anyone else be running multiple isolates
272.
by mrkurt 4y ago
Or a proxy.
273.
by mrkurt 4y ago
Oh I see! Fly Machines are lower level than what most people use us for. They're designed for people building platforms. Most people run "apps" on Fly. You control which regions they run in, we load balance to the nearest. We
274.
by mrkurt 4y ago
This is separate from the Nomad stack. When you run `fly launch` you get a Fly App that's orchestrated by Nomad and manages something-like-fly-machines for you. Fly Machines are nearly orchestration free. Machines are designed to work
275.
by mrkurt 4y ago
Some of the plumbing is similar, but in general: 1. Lambda has runtime constraints and is billed per request 2. Fly Machines are basically just VMs with some magical startup sugar. You can do pretty much anything you want with them, includi
276.
by mrkurt 4y ago
That is talking about our S3 buckets, when you use these you don't know you're using S3. Fargate and Lambda are not as consistently fast to boot VMs. Fargate, in particular, can take minutes to get a container launched. This is
277.
by mrkurt 4y ago
We do not. I'd like to, but we have a lot more infrastructure work to do before it's even possible. At the moment, we don't even migrate persistent disks or IP addresses between hosts.
278.
by mrkurt 4y ago
It implies that they didn't succeed.
279.
Phoenix and LiveView with Chris McCord(frameworkfriends.com)
3 points by mrkurt 4y ago|0 comments
280.
by mrkurt 4y ago
This pretty much fits our definition of "deploy app servers close to your users". One region works just fine for some apps. Some are worth going to three.
281.
by mrkurt 4y ago
$0.9 is a typo, I meant $0.09/GB in Chennai. We're not taking that much of a haircut.
282.
by mrkurt 4y ago
This is part of why the PaaS take has worked so well for us. People who think they want edge have all kinds of different needs. When we realized that all full stack devs could benefit from something kind-of-like-edge it helped us do more
283.
by mrkurt 4y ago
"We also waive fees from big, legitimate bursts."
284.
by mrkurt 4y ago
No, we don't bill people for traffic from attacks. We also waive fees from big, legitimate bursts. The intent of our bandwidth pricing is to allow high usage, sustained bandwidth workloads. It's not to sneak one over on you.
285.
by mrkurt 4y ago
This is embarrassingly non obvious in the docs, but you can run workers/web just like you would on Heroku: https://community.fly.io/t/preview-multi-process-apps-get-yo... Most people run workers in their primary r
286.
by mrkurt 4y ago
This is pretty much what I believe. This isn't HN frontpage worthy, but one of the things I'm most excited about is people running production CockroachDB clusters on Fly.io. It is still a little more difficult to use the underlyin
287.
by mrkurt 4y ago
Don't tell anyone, but you can try "fly machines run" (currently only in a new, empty app). Then "fly machine start <id>" to start the thing back up after it exits. It's not released yet, though.
288.
by mrkurt 4y ago
There are three ways to manage bandwidth prices: 1. Put servers where bandwidth is cheap (not Sydney, for example) 2. Constrain throughput per server 3. Buy from cheap transit providers like Cogent Hetzner does all three. Bandwidth in the U
289.
by mrkurt 4y ago
Database read request are not the same as readonly HTTP requests. I am much happier having all requests hit my app process than I am trying to do the CDN dance. Right now your choices are: run a database in on region and: 1. Use the weird H
290.
by mrkurt 4y ago
It's very generous of you to assume we're focusing on underlying platform. It's true! But it's a difficult thing to notice when our service gave you papercuts. We've just now grown large enough to have people focus
291.
by mrkurt 4y ago
We actually launched with that feature: https://news.ycombinator.com/item?id=22616857 No one took us up on it. What we found is that the majority of people want their stuff to stay up, and the right UX for "shut it dow
292.
by mrkurt 4y ago
This is the distribute-your-Rails-app-without-making-any-code-changes version of that story. It works great for apps that are 51% or more read heavy. You drop our library in, add a region, and off you go. The library takes care of eventual
293.
by mrkurt 4y ago
Oh but you are! And it won't even cost you anything. I'd bet money your blog fits in our free tier, we just don't (a) tell you that and (b) solve the "what happens if there's a bandwidth burst" problem.
294.
by mrkurt 4y ago
This is a problem. And a bit of an own goal on our part. I hate services that don't put a price on things like bandwidth (because there's always a price!). So we priced bandwidth and made it transparent. You can put an app on Fly.
295.
by mrkurt 4y ago
We are DigitalOcean circa 2015 mature. Good enough for some very large customers, good enough for many developers, not some place I'd run pacemaker infrastructure.
296.
by mrkurt 4y ago
First, Heroku's pipelines and GitHub integration are (or, I guess, were) excellent. We (Fly.io) intentionally didn't build a pipeline replacement. We exist for one reason: to run apps close to users. We're just now to the siz
297.
by mrkurt 4y ago
We've been kicking around ideas for managing files on volumes. This is a common problem – it's actually more difficult than you'd expect because "securitah". Once your volume is mounted in one of your VMs, we can&#x
298.
by mrkurt 4y ago
You are correct about Laravel apps, but there's some nuance here. The Heroku magic only worked for Rails in the early days. We (Fly.io) have a similar experience for Phoenix, Rails and Remix apps. We don't have magic for Laravel y
299.
by mrkurt 4y ago
that function is correct when it agrees with me.
300.
by mrkurt 4y ago
We have been chasing this down for weeks and can't find the actual bug/workaround here. It's definitely IPv6 related, we think having something to do with weird MTUs. Are you using an IPv6 tunnel or connecting via a vpn by ch

More ›