One of the common things that happen when folks just land in Fly.io is that it’s not ultra clear when their app is misbehaving. Its always clear when a Build or a Deployment fails because, well, it says so. But what happens when after deploying stuff, stuff is actually not working properly?
We recently shipped a new helper to dashboard App UIs to be very clear when something is not doing well in your app. We call it Fly Doctor (shoutout to anyone who used the CLI version).
It will read some primitives like proxy logs and machine state on your behalf and infer ongoing issues and will show them clearly on your app dashboard.
Alongside it the existing Logs & Errors page will come with some tips on how to debug your existing issue. Those are the very same hints we always bring on docs and here on community!
What’s next
Right now Fly Doctor is on its infancy, it only handles a few cases related to the most common error symptoms in newly-launched app such as proxy issues and machine restarts (trust me, these already cover a lot!). But we do and will ship more scenarios here.
What about you? What you wish Fly.io told you the very first time you deployed something?
I checked fly-proxy logs for your organization and there’s some concerning logs as recently as 4mins ago:
[PR04] could not find a good candidate within 20 attempts at load balancing
and
[PC01] instance refused connection. is your app listening on 0.0.0.0:3000? make sure it is not only listening on 127.0.0.1 (hint: look at your startup logs, servers often print the address they are listening on)
Maybe your app had slow start after a deployment and should be good to go now.
Having the same message as @simonh despite no issues connecting to the app, persisting through machine restarts (with no errors in the logs and the restart happening quickly)
