Health checks suddenly started failing from Tuesday

I deployed my code on Tuesday morning, it was fine. I then wanted to redeploy the same code with a minor change and healthchecks on flyio were failing.

I went back to the commit that worked and tried to deploy that, still failing health checks.

I can see in the logs that the server boots up to port 3000 just fine. If i build and run the build locally i can access the health check endpoint which is “/heartbeat”.

The logs show this:

2025-12-17 12:08:31.924 Health check ‘servicecheck-00-http-3000’ on port 3000 is in a ‘warning’ state. Your app may not be responding properly. Services exposed on ports [80, 443] may have intermittent failures until the health check passes.
2025-12-17 12:08:31.618 2025/12/17 10:08:31 INFO SSH listening listen_address=[fdaa:c:5e10:a7b:1cf:a63f:5d39:2]:22
2025-12-17 12:08:31.427 Machine created and started in 15.747s
undefined -— -—
2025-12-17 12:08:31.286 INFO [fly api proxy] listening at /.fly/api
2025-12-17 12:08:31.282 INFO Preparing to run: `docker-entrypoint.sh node dist/src/server.js` as root

I cant seem to get deployments to work, i have been deploying to fly for over a year so I am not new to deployment issues, the config or cli and this issue has stumped me.

It looks like fly has changed something their side because the health check happens prior to fly setting up the proxy. Im not sure how the health check is done their side.

Even on my production machines that are running and serving users, their health checks are showing red 0/1 as well. Thank goodness they are still running.

Seems like the flyio health check service has been damaged by some change and is now blocking new deployments.

Hey @greg-cbs

Thanks for the report.

That was a bug. New state synchronization code was enabled in jnb yesterday and it had a bug in health checks handling.

Should be fixed now.

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