[ruby-core:108206] [Ruby master Feature#17849] Fix Timeout.timeout so that it can be used in threaded Web servers

Issue #17849 has been updated by avit (Andrew Vit).


The handling of timeouts is [documented][1], but that only shows it with a `handle_interrupt` block set up around everything, not under `ensure`. In the examples shown here, can `ensure;begin` or `ensure;Thread.handle_interrupt` be considered "atomic", or is it possible for an interrupt to happen between these lines?

(Another rubyist once recommended a [different approach using rescue/retry][2], but in my testing it looks like `handle_interrupt` is more robust.)

I have a small thought on a possible API: if it might be an improvement to pass interrupt handling options to `ensure`, similar to `rescue`:

```
begin
ensure Timeout::Error => :never
  $finalized = true
end
```

To extend the similarity with `rescue`, could it make sense to allow multiple ensure blocks, if different options are needed in sequence?


[1]: https://ruby-doc.org/core-3.1.1/Thread.html#method-c-handle_interrupt-label-Guarding+from+Timeout-3A-3AError
[2]: https://web.archive.org/web/20110113205935/http://blog.segment7.net:80/articles/2006/04/11/care-and-feeding-of-timeout-timeout


----------------------------------------
Feature #17849: Fix Timeout.timeout so that it can be used in threaded Web servers
https://bugs.ruby-lang.org/issues/17849#change-97190

* Author: duerst (Martin D端rst)
* Status: Open
* Priority: Normal
* Assignee: matz (Yukihiro Matsumoto)
----------------------------------------
Making this a separate issue from #17837

Eregon (Benoit Daloze) wrote in https://bugs.ruby-lang.org/issues/17837#note-10 (which is  about timeouts for regular expressions):

> I think fixing Timeout.timeout might be possible.
> The main/major issue is it can trigger within `ensure`, right? Is there anything else?
> We could automatically mask `Thread#raise` within `ensure` so it only happens after the `ensure` body completes.
> And we could still have a larger "hard timeout" if an `ensure` takes way too long (shouldn't happen, but one cannot be sure).
> I recall discussing this with @schneems some time ago on Twitter.





-- 
https://bugs.ruby-lang.org/

Unsubscribe: <mailto:[e-mail]?subject=unsubscribe>
<http://lists.ruby-lang.org/cgi-bin/mailman/options/ruby-core>