Return-Path: william@bourbon.usc.edu
Delivery-Date: Fri Sep 26 08:05:08 2008
X-Spam-Checker-Version: SpamAssassin 3.2.3 (2007-08-08) on merlot.usc.edu
X-Spam-Level: 
X-Spam-Status: No, score=-2.3 required=5.0 tests=AWL,BAYES_00 autolearn=ham
	version=3.2.3
Received: from bourbon.usc.edu (bourbon.usc.edu [128.125.9.75])
	by merlot.usc.edu (8.14.1/8.14.1) with ESMTP id m8QF58in028856
	for <cs551@merlot.usc.edu>; Fri, 26 Sep 2008 08:05:08 -0700
Received: from bourbon.usc.edu (localhost.localdomain [127.0.0.1])
	by bourbon.usc.edu (8.14.2/8.14.1) with ESMTP id m8QF8TRS009359
	for <cs551@merlot>; Fri, 26 Sep 2008 08:08:29 -0700
Message-Id: <200809261508.m8QF8TRS009359@bourbon.usc.edu>
To: cs551@merlot.usc.edu
Subject: Re: Regarding ctrl+C
Date: Fri, 26 Sep 2008 08:08:29 -0700
From: Bill Cheng <william@bourbon.usc.edu>

Someone wrote:

  > I dont know if this is a proper solution to the problem, but I have a main
  > thread which spawns the arrival thread and S1 and S2. I set the main thread
  > to capture and handle the signal. Since the main thread never goes to sleep,
  > i believe that the signal will not be lost. Is this OK?

I cannot answer the "is this okay" type questions.  Please
see:

    http://merlot.usc.edu/cs551-f08/description.html#email

There is really not much difference whether you use the
main thread or the arrival thread to catch <Cntrl+C>.
Your main thread *must effectively also sleep*, i.e., get
blocked waiting for the <Cntrl+C>!  If you do busy-wait,
you will lose 30 points!
--
Bill Cheng // bill.cheng@usc.edu <URL:http://merlot.usc.edu/william/usc/>





  On Thu, Sep 25, 2008 at 10:15 PM, Bill Cheng <william@bourbon.usc.edu>wrote:
  
  > Someone wrote:
  >
  >  > I have a question about the signal handling. Suppose my arrival thread
  >  > has already put 2 customer in the customer queue. The arrival thread is
  >  > currently sleeping for fill up the interarrival time of the 3rd
  >  > customer. Suppose the interarrival of the 3rd customer is 10secs. Thus
  >  > arrival thread sleeps for 10 secs, so it will go in waiting processes
  >  > queue. In this time, ont of the server threads will run. What if I press
  >  > ctrl C in this time? The arrival thread who was responsible for handling
  >  > this signal will not catch the signal (as it is on waiting queue).
  >  > Should we handle ctrl C in S1 and S2 also? This problem doesn't happen
  >  > frequently, but it can be seen in sample trace file when ctrl C is
  >  > pressed after 11secs.
  >
  > If you call select() or usleep() to sleep in your arrival thread,
  > select() or usleep() should return with an error condition and
  > errno should be set to EINTR.  If this is not happening, may
  > be you did not setup the sigmask correctly.  Please do:
  >
  >    man pthread_sigmask
  >
  > to see how to catch <Cntrl+C> in pthread.
  > --
  > Bill Cheng // bill.cheng@usc.edu <URL:http://merlot.usc.edu/william/usc/>