Subject Re: [ib-support] Re: --* Please Help*--- Database corruption and cannot backup & restore
Author Paul Schmidt
Lester:

On 9 Aug 2001, at 10:16, lester@... wrote:

Organization: L.S.Caine Electronic Services
To: ib-support@yahoogroups.com
From: lester@...
Date sent: Thu, 09 Aug 2001 10:16:05 +0100
Send reply to: ib-support@yahoogroups.com
Subject: Re: [ib-support] Re: --* Please Help*--- Database corruption and cannot backup & restore

> > I've been developing with W2kpro for more that a year now, and I
> > never had to reboot the machine other than shutting it down at the
> > end of the day when I go home. I must have been lucky! <g>
>
> I think you have provided your own answer - shutting it down at the
> end of the day.
>
> Do that and I don't think you have been lucky, you are clearing the
> faults every day. Try leaving it on for a week while you soak test
> something - I have a live feed of train movements feeding the soak
> system.. ( Remember that Win98 crashed after 43 days - I hit that one
> - Microsoft deniged there was a problem - then fixed it! ).
>
> My system never gets switched off - which is the problem.
>

In the case of servers, if you absolutely need to run Windows, you
need to schedule regular reboots, I think there is a utility that
will allow you to set up a reboot as an AT job. If you can't reboot
frequently then you need something else. Most Unixes are good, they
can often run years between reboots, although the occassional reboot
is good for any machine.

In the case of Workstations or stand alones, they should be "rested"
at the end of the day, if for nothing else then the fact it reduces
energy usage. 20 years ago when computers were expensive,fragile
things and energy was relatively cheap, it made sense to let machines
run, because the cost of replacing components more often, outweighed
the few dollars per year that it used in energy. Now a machine that
runs 7/24 can use up more energy in 2-3 years then it costs to
replace the entire computer, which is robust enough that obsolecence
will kill it, long before the on/off cycle of the components could
ever get it.

Paul

Paul Schmidt,
Tricat Technologies
Email: paul@...
Website: www.tricattechnologies.com