←back to thread

528 points sealeck | 1 comments | | HN request time: 0s | source
Show context
sergiomattei ◴[] No.31390808[source]
I don’t feel that way at all.

Every time I’ve tried Fly (trust me I’ve wanted to love it), there’s always a rough edge or the service breaks for me.

First time I tried it, the web panel wasn’t even loading. Second time, months later, everything was 500ing and I couldn’t find a way to SFTP into a disk (!!!). Total dealbreaker.

This was easily done in Render.com with an even more magical experience. Deploy from a GitHub repo and I was live in minutes. Upload the files from local and done.

I want to love Fly so much. I align with their mission. I love their first class Elixir support. But so far I’m not impressed.

It looks to me like Render is seriously taking the PaaS crown at the moment, with innovation after innovation, affordable pricing and excellent user experience.

replies(2): >>31390822 #>>31390891 #
mrkurt ◴[] No.31390891[source]
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't run tools outside the VM to let you manage the file system. On something like k8s with vanilla Docker, we could. But no one should run multitenant Docker.

It's not really an excuse, just a reason it's taking longer to solve than we'd like.

The errors on the web UI sucked. These have improved drastically in the last three months (because we have smart, dedicated people working on fullstack for us now).

replies(2): >>31390901 #>>31392117 #
joseph ◴[] No.31392117[source]
> 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't run tools outside the VM to let you manage the file system. On something like k8s with vanilla Docker, we could. But no one should run multitenant Docker.

Was there a reason for not using something similar to kata containers where you run a microvm but still use containers inside them? It seems like it would make such things easier while getting the isolation of a VM.

replies(1): >>31395385 #
jpgvm ◴[] No.31395385[source]
They do that already AFAICT by using Firecracker.
replies(1): >>31395667 #
1. joseph ◴[] No.31395667[source]
Unless they’ve changed things, there is no containerization within the VM a la kata. They run their own custom init inside the VM and use it to start the entry point. https://github.com/superfly/init-snapshot is the source.