Path: utzoo!utgpu!water!watmath!clyde!att!osu-cis!tut.cis.ohio-state.edu!rutgers!apple!bionet!agate!labrea!decwrl!sun!pitstop!sundc!seismo!uunet!mcvax!ukc!etive!hwcs!zen!frank
From: frank@zen.UUCP (Frank Wales)
Newsgroups: comp.misc
Subject: Re: computer follies
Summary: some do
Message-ID: <1417@zen.UUCP>
Date: 11 Oct 88 12:11:58 GMT
References: <916@viscous> <1086@bucket.UUCP> <23134@amdcad.AMD.COM> <4041@tekgvs.GVS.TEK.COM>
Reply-To: frank@zen.co.uk (Frank Wales)
Organization: Zengrange Limited, Leeds, England
Lines: 27

In article <4041@tekgvs.GVS.TEK.COM> toma@tekgvs.GVS.TEK.COM (Tom Almy) writes:
>In article <23134@amdcad.AMD.COM> rpw3@amdcad.UUCP (Rob Warnock) writes:
>>On PDP-11's with real core (not this modern RAM stuff!), there was a
>>power-fail interrupt that gave you enough time to save critical state
>>and go into a loop before the CPU quit working.
>
>I was once running a several hour long BASIC program
>when the power failed.  Sure enough, when power was restored, the program
>continued to successful completion.  
>
>I'd sure like to see systems do that these days!  (Especially a *NIX system!).

Well...when power failed to our building a few months ago, it took about
a quarter of a minute for our standby generator to kick in and start
supplying enough juice to get things going again.  At that point,
our HP 9000/840 restarted, waited for all the disks to get back up to
speed, synced them, checked them, then carried on as if nothing
had happened.  Battery-backed RAM [16MB] may not be quite the same as core,
but it ain't half useful sometimes, especially when the OS supports it.
[HP-UX!  Ra, ra, ra!  HP-UX!  Ra, ra, ra!]

(Now if only the terminals could have preserved the contents of their
screens too...)

--
Frank Wales, Systems Administrator,  [frank@zen.co.uk<->mcvax!zen.co.uk!frank]
Zengrange Ltd., Greenfield Rd., Leeds, ENGLAND, LS9 8DB. (+44) 532 489048 x220