Hi,
A bit of an odd one that I can’t really point the finger to yet, but maybe someone has some insights that could help me trace this.
I’ve got a running app that listens on a TCP port. The app was assigned a dedicated IP (v4). The app is available, working and responsive, so the issue isn’t the app itself. However, I’m having some difficulties on connecting from certain network environments.
From my laptop, I can connect and the link is stable. Bi-directional communication works as expected. However, I have an IoT device that is failing to establish a connection to this address.
If running the application locally on my machine, proxying the TCP with ngrok, the device connects and communicates properly. So it’s not the app/protocol.
If I put the application somewhere else, let’s say, Azure Containers, with a public IP and an exposed port, the device connects and communicates properly.
However, the application hosted on fly, the device fails to connect; unfortunately I haven’t managed to get my hands on the logs from the device, since it’s “out there”, but will do.
At first, I though the DNS might not be resolving, so I used the IP address instead. But the results are the same…
The only thing different that I can see, is the network. The device is using a mobile network to communicate, plus, it’s on roaming mode (not sure if relevant). Would it be possible for the IP to not be reachable “partially”?
I’m posting this here, since the network environment seems to be the difference. As I said, if hosted somewhere else with a public IP and port, using the same image, there are no issues.
If anyone has any suggestions… I’d appreciate it.