Heya, seems your server is taking too long and we consider the app as failing to respond even though it's booting. I've raised this internally to see how we can improve this.

thanks !

The maximum time we wait for is 10s. If you need to run long functions at the start of your service (ex: pulling data, etc), you should prep and cache those in a predeploy step.

I'm having the same problem as well. The first fetch() request is simply failing with Unable to connect. Is the computer able to access the url?. I am under the Railway's Private Networking to connect to this particular service. Is there perhaps a bug where Serverless doesn't work with Private Networking? The failure happens faster than 10s. It's nearly immediate.

You would need to implement retries in your code for communication privately with services that go to sleep, while our public proxy will hold the request for up to 10 seconds, the same does not happen over the private network, so it is up to your code to handle that.