From: Paul Brannan <pbrannan@...>
Date: 2004-09-28T23:10:06+09:00
Subject: Re: safety of timeout()

On Tue, Sep 28, 2004 at 08:47:29PM +0900, Brian Candler wrote:
> At least, that's what I understand to be the crux of the issue. In many
> cases it's not going to be a major concern. You may be able to rewrite the
> code to make it safer by pushing the timeouts down to the lowest possible
> level. The above example could be rewritten safely as:
> 
> File.open("mylog","a") do |f|
>   timeout(30) do
>     ... do stuff
>   end
> end

I like your description of the problem, Brian.  I've tried to explain
this before and not been able to articulate it quite so well.

I would like to add one additional problem to your explanation, though:
timeout exceptions can prevent you from knowing whether or not an
operation actually succeeded.  We have this problem with CORBA timeouts
in particular; a remote call is made, and the process on the other side
is running slowly.  It does, however, complete the operation, just as
the timeout exception is being fired.  So do I treat the operation as
success (since I know the request reached the other side, as I didn't
get a communications failure exception), or do I treat the operation as
failure (since I did get an exception and I did fail to get a return
value from the call)?

A solution I've used in the past has been to use an event loop and let
it handle the timeouts.  The timeout then occurs only when the event
loop has control; when the timeout does occur, a proc is called that
handles the timeout.  This proc may raise an exception or it may take
some other action.

For long-running operations, I periodically yield control to the event
loop when it is safe.

Paul