What should happen if deliver_cancel raises an exception? In the current implementation, the shielded cancel scope will prevent it from propagating until the process exits, which might take a while if the crash occurred before signaling the process in any way. Maybe on exception from a user-specified deliver_cancel we should call the default deliver_cancel to kill the process? Or just kill() since

Steps to reproduce

• futures-preview = { version = "0.3.0-alpha.19", features = [ "async-await" ] } in Cargo.toml as recommended in §3.1.
• Try to build the example in https://book.async.rs/tutorial/sending_messages.html

use futures::channel::mpsc; // 1
use futures::sink::SinkExt;
use std::sync::Arc;

type Sender<T> = mpsc::UnboundedSender<T>; // 2
type Receiver<T> = 

We had an issue where an api user was using the incorrect parameters on an API call. Having this feature would have made the API fail instead of proceeding. The bug would been found earlier.

I can see where this would be a hassle on many cases, But it would be nice for new projects.

I suppose it could be a global config item that could be overridden at the individual action level:

For