Requests are getting stuck during network connection drops

Describe the bug

I've observed that, specifically in scenarios where there is network instability and traffic cannot be routed for some amount of time, instead of receiving an exception, requests appear to get stuck. On the other hand, when I forcefully shut down a vCSA VM (for instance), I receive a proper exception that can be handled within our code, something along the lines of a ConnectionError.

Reproduction steps

1. Establish a connection with the vCSA VM using SmartConnect
2. Call any API in an infinite loop
3. Intentionally cause a break in the network without actually bringing down either the client or the server
4. The API call seems to get stuck, and does not resume even once the connection has been restored

Expected behavior

As is occurring with the shut down of the vCSA VM, in this scenario I would expect an exception to be thrown and for the request session to be closed.

Additional context

If this is either a known issue or if there's some kind of exception handling that I'm overlooking, please do let me know.

Start with the SmartConnect entry point and reproduce the reported API call while interrupting network traffic without stopping either endpoint. Trace the request and session handling during the interruption; done means the call raises a handled connection exception and the request session is closed rather than remaining stuck.