Function to kill a running process

[hayd]

There doesn't seem to be a way to kill a process initiated by run.
close only removes it from the resources list.
A running process continues after the original deno application is complete.

A few options:

1. Kill all running/open processes when deno is exited
2. Close should kill the process.
3. There should be a new function to kill a process.

I favor 1+2.

[ry]

Close ought to kill it. And #1 should be true too. I think these are expected behavior.
Thanks for reporting.

[hayd]

1. Doesn't appear to be the case, example:

import { run } from "deno";

const justAMinute = run({
  args: ["python", "-c", "from time import sleep; sleep(60);"]
});

(async () => {
  // 5 seconds
  await new Promise(res => setTimeout(res, 5000));
  // justAMinute.close();
})();

// python is still running

This stemmed from this close in the deno_std test not cleaning up the file_server. Which I can also still replicate.... Could this be that the .close is not waiting for the response? It's weird as in the example above adding the close does work as expected.
If I add an await for a second after this line (in test.ts) it does clean up for me.

Note/Aside: If you kill -9 an application, e.g. python, they usually won't kill of their child processes (I assume this is a unix standard)...

[hayd]

I wonder if std::process::exit is the issue and we're not calling all the destructors.

Note that because this function never returns, and that it terminates the process, no destructors on the current stack or any other thread's stack will be run. If a clean shutdown is needed it is recommended to only call this function at a known point where there are no more destructors left to run.

From the documentation it reads like this ought to be dropped (and closed).

---

This seems similar to #1517 cc @kevinkassimo ....

[ry]

It’s likely that we need to create a “process group” and stick all child processes in there (it’s a posix thing).

[kevinkassimo]

@hayd I'll also investigate. Thanks for pinging me

[bartlomieju]

@hayd any update on this?

[kevinkassimo]

Sorry... Working on this (seems easy enough to fix)

[kevinkassimo]

Hmm I have just tested, it seems that the example @hayd provides is working now with latest release...