[BUG] host shutdown with a Ctrl-C exits with a warning

[protochron]

Affected project(s)

• [ ] documentation
• [ ] examples
• [X] wasmCloud host
• [ ] wasmCloud CLI (wash)
• [ ] wasmCloud dashboard UI (washboard)
• [ ] capability providers
• [ ] provider bindgen
• [ ] control interface client
• [ ] other / not sure

Describe the bug

Stopping the host with Ctrl-C successfully kills the host, but it ends with a warning that the wasmtime epoch interrupt thread is timing out:

Error: failed to shutdown host

Caused by:
    epoch interrupt thread timed out

Steps to reproduce

1. Start a wasmcloud host, wadm, nats
2. Start a wadm manifest (I've been using the rust http hello app)
3. Ctrl-C in a terminal to stop the host
4. Observe the error

Expected behavior

I expected the host to stop without that warning.

If it's a bug in the way we're stopping wasmtime the we should fix it, otherwise we should downgrade the warning if there is nothing for us to do.

Environment

• OS: OSX
• Shell zsh
• wasmcloud Version: v1.3.0

Screenshots / Logs / Additional context

No response

[rvolosatovs]

It's actually a duplicate of https://github.com/wasmCloud/wasmCloud/issues/2391, which was closed a few months ago. Not sure if it's a regression though, I think this is simply a race condition in shutdown sequence, which was never fixed.

I also encounter this one quite often

[brooksmtownsend]

Yep I think this isn't a regression as much as it just wasn't truly resolved 😅

[brooksmtownsend]

Digging in, this is almost certainly due to any running component taking a reference to the underlying Runtime Engine and the epoch thread not exiting until every reference is dropped.