Date: Tue, 28 Jul 1998 11:11:16 +0300 (IDT)
From: Eli Zaretskii <eliz AT is DOT elta DOT co DOT il>
To: Nate Eldredge <nate AT cartsys DOT com>
cc: Martin Str|mberg <ams AT ludd DOT luth DOT se>,
        DJGPP-WORKERS <djgpp-workers AT delorie DOT com>
Subject: Re: A call to `sync' inside `spawn'
In-Reply-To: <35BD5A3D.3B5C56E1@cartsys.com>
Message-ID: <Pine.SUN.3.91.980728110724.4133O-100000@is>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Precedence: bulk


On Mon, 27 Jul 1998, Nate Eldredge wrote:

> > And sync on my machine (Linux) doesn't take several seconds. Less than
> > one, actually, and it's not idle.
> 
> I think it will depend on what the system is doing.  On Unix, it's
> traditional for the `update' daemon to call `sync' every 30 seconds or
> so, so if it's been fairly idle since then, there won't be much to do. 

I don't want this thread to become Unix sysadmin class, but 
``traditional'' is too optimistic.  I have seen Unix machines that don't 
sync at all.  In addition, some varieties of Unix can produce a lot dirty 
buffers in the cache, depending on what they are doing.  One particular 
case that gave me trouble was consistently wiping out files on power 
outage, even though they occured many minutes after a file was created.  
(I solved that by forcing a sync in the Makefile that produced the 
files.)