HP3000-L Archives

August 2002, Week 4

HP3000-L@RAVEN.UTC.EDU

Options: Use Monospaced Font
Show Text Part by Default
Show All Mail Headers

Message: [<< First] [< Prev] [Next >] [Last >>]
Topic: [<< First] [< Prev] [Next >] [Last >>]
Author: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
Subject:
From:
"Wayne R. Boyer" <[log in to unmask]>
Reply To:
Date:
Thu, 22 Aug 2002 13:12:53 EDT
Content-Type:
text/plain
Parts/Attachments:
text/plain (29 lines)
In a message dated 8/22/02 12:29:23 AM Pacific Daylight Time,
[log in to unmask] writes:


> There's one problem with this - I have yet to see a terminal or PC that can
> disallow a pulled plug. Any kind of power hit is going to produce a rude
> disconnect that the server simply cannot resist.
>
>

This whole discussion is a very good one.  My opinion is that however
possible, software should be designed to encourage orderly exiting and
termination of transactions.  As such, I think that this proposed enhancement
to QCTerm is a good request.  By disallowing at least one form of 'bad' user
behavior, some problems can be prevented.  Not all, but some.

Nothing in software is going to be able to handle a yanked power plug (UPS
units for every desk?), a Windoze crash/freeze (never happens...), a
cut/pulled cable, etc, etc, and more etc. Hopefully these are all unplanned
accidents and somehow, application software should be designed so that no
data corruption can occur at any point in any transaction.  Many if not most
applications just are not that robust though.

Wayne Boyer
Cal-Logic

* To join/leave the list, search archives, change list settings, *
* etc., please visit http://raven.utc.edu/archives/hp3000-l.html *

ATOM RSS1 RSS2