could not make HTTP request to instance

We keep receiving the following errors in our monitoring every few hours.
We have 3 machines running in a region, and two of them throw these errors at the same time.
Then the load goes to third machine and our graphs show a high response time.

We have not set up any healthchecks on our app and we are pretty sure that the app is solid, as we hav been running it on other platforms oreviously.

What is the real cause of this error, is it something in our app or is it the fly.io’s servers are crashing?

The app is actually deployed in 4 regions but the issues appears only in singapore region.


2023-07-28T14:50:43.629 proxy[xxxxxxxxxx] sin [error] could not make HTTP request to instance: connection error: timed out

2023-07-28T14:50:59.496 proxy[xxxxxxxxxx] sin [error] could not make HTTP request to instance: connection error: timed out

That error means out proxy tried to do a HTTP request to one route of yours and it took so long that it timeouted, I had that same issue on an app of mine a few weeks ago.

My recommendation is try to pinpoint which endpoint is slow and see if it’s an intermittent issue there.

This topic was automatically closed 7 days after the last reply. New replies are no longer allowed.