I'm a pretty happy Fly.io customer and glad to see the direction things have been going overall. While it's great to see Fly is getting more resources to continue improving and building the business I worry about the inevitable VC Countdown Clock to Exit.

What does this new round mean for Fly's long term independence?

I don't think we can build a company like this without a lot of money.

We're to the point where we're making $10-25mm capital expenditures in one go. And we have to to optimize our own costs. Optimizing our own costs means long term independence.

What I think I've learned about VCs is: the trick to long term independence is to find investors who already value the sales mechanism. Bottoms up, dev focused "sales" is something investors like, now (this was not true in 2012).

Which gem? There are several that bind against libcurl.

I use Typhoeus which uses Ethon depends on ffi. Crash happens. Sorry English is not my mother language.

I'll have to check when I get into work.

I didn't realize curb was still maintained. Would be helpful to report the issue here: https://github.com/taf2/curb/issues

Yeah, I was/am planning to do that. But... the project giving us trouble is still on 0.9.11 and another project that is on latest doesn't seem to be having the same issue, so some more testing is needed. At the moment, we're going to rebuild using the older base image and then we'll upgrade the gem and test this sprint.

Update: Latest gem still has the elevated error rates from the service we're connecting to but NOT the resource leak.

We recently swapped our metrics to Grafana Cloud and were really surprised (despite being documented) that pricing is based on samples per minute not metrics series. So, for example, if we send a metric every 15s (the Prometheus default) then we get charged as if that were four separate metrics. Support was very helpful explaining everything and they reversed the charge but it still feels weird.

Thanks for the feedback! Agree that this can be made more clear and we will work on that.

We experienced the same thing. Pulled the plug on using the cloud version

Close but not quite.

1 is always free (includes email notifications)

$5/month for 3 with email notifications

$19/month adds integrations for notifications through services like Slack, PagerDuty, and Squadcast (among others).

Ah. Thank you for the clarification. Still a bit pricey for what it is, but that's very nice having a free tier.

Hopefully what you’re monitoring is valuable to you or the company you’re working for (otherwise why monitor it?).

If you have a personal project and need a few extra monitors (Snitches) then reach out to support. Happy to help out any way we can.

Simple and reliable take a surprising amount of time to get right. We would rather move slowly and cautiously because we have thousands of operations folks who could have a really bad time if we rush something that causes folks to get paged at 2am or lose trust in the system.

Also, we’re bad at talking about the work we do. It’s the classic trap a lot of engineering founders fall into.

Lead developer behind Dead Man’s Snitch here :wave:

We are definitely not in “maintenance mode” and have a lot of things in the works. With the last couple of years being what they’ve been things have often taken longer than we’d like but they’re certainly in motion.

edit: fixed tpyo

> Lead developer behind Dead Man’s Snitch here :wave:

Class! ;)

Great to hear. Have used Dead Man's Snitch in the past and plan to do so again in the future.