[clug] Microsoft Azure down for 9 hours with " leap-year date (problem) that prevented systems from calculating the correct time."

Rhys Ambler rhys at ambler.id.au
Fri Mar 2 04:14:01 MST 2012


Microsoft still supply something that is not as well recognised in
opensource: a scape goat.

I will always go back to the example where I was told not to use 7zip
because the license meant that if/when something went wrong we had no
recourse.
Where there is a corporate mentality that pointing the finger at someone
else for making a problem somehow alleviates the problem then free open
source will never gain traction. It is hard enough for supported opensource
to gain traction while there is a mentality that Microsoft is the serious
option for business.
Those in the know in IT may disagree, but among so many people I meet the
Wintel world is the norm and apple is the only real alternative.
As long as these people are the ones who sign off on purchases the Wintel
world is secure.

It is scary to see how reliant we are becoming on such delicate
infrastructure.

Rhys Ambler
+61 417 289 103
rhys at ambler.id.au



On 2 March 2012 16:01, steve jenkin <sjenkin at canb.auug.org.au> wrote:

> Two articles by Dan Goodin with same text, different photos.
>
> <
> http://arstechnica.com/business/news/2012/03/how-a-leap-year-bug-could-topple-microsofts-azure-cloud-service.ars
> >
> <http://www.wired.com/wiredenterprise/2012/03/azure-leap-year-bug/>
>
> and a comment from a Gartner 'Cloud Expert':
>
> <
> http://blogs.gartner.com/kyle-hilgendorf/2012/02/29/another-cloud-outage-insight-and-reactions/
> >
>
> And the Rest of the World, based on POSIX, seemed to sail on regardless...
> Linux and OS/X [and AIX, Solaris, HP/UX, ...] Just Work in this.
>
> Getting something so basic wrong, something that has been definitively
> solved for over half the lifetime of Commercial Computing (!), says
> there are serious flaws in both the Microsoft Software Development and
> Acceptance processes *and* their basic competence/capability in writing
> Software...
>
> Why don't commentators draw attention to this???
>
> I can't remember the full history of Time Zones in Unix.
> Don't recall them being there in V6 or V7. UNSW & early BSD had them
> from memory. [Please correct me!]
>
> Seem to recall that Adelaide, being 0:30 behind Aus East Coast, caused
> the first attempt with whole hours to be redone (TZ="5EDT6", where time
> behind GMT/UTC was positive and for Australia (ahead) TZ="-10EDT-11",
> IIRC.)
>
> Full Internationalisation was definitely there with System V Release 3
> (SVR3 included streams and "file system switch" as well). Was that same
> time as BSD 4.3/4.4 or a little before?
>
> Previously, by System V 2.2, there were two competing 'curses' libraries
> and databases (SysV - terminfo, BSD - termcap [Bill Joy]).
> There were so many ascii terminal makers around that 'terminal
> definitions' couldn't be hard-coded.
>
> Unix wasn't born being 'portable' and 'the Universal glue', but that
> followed on so fast as to be indistinguishable to most folk.
>
> Not sure how long it took before we had "8-bit clean" utilities in Unix.
> SVR4?? [ie. a 'null' character didn't stop reading, and not just 7-bit
> ANSI/ASCII characters supported]
>
> It took Ken and Rob Pike to come up with UTF-8, and create a system
> (Plan 9) that began life without any 7- bit ascii limits, ie native
> UTF-8 and UTF-16.
>
> --
> Steve Jenkin, Info Tech, Systems and Design Specialist.
> 0412 786 915 (+61 412 786 915)
> PO Box 48, Kippax ACT 2615, AUSTRALIA
>
> sjenkin at canb.auug.org.au http://members.tip.net.au/~sjenkin
>


More information about the linux mailing list