[154273] in North American Network Operators' Group

home help back first fref pref prev next nref lref last post

Re: FYI Netflix is down

daemon@ATHENA.MIT.EDU (Jimmy Hess)
Sat Jun 30 15:28:40 2012

In-Reply-To: <CAB2RJygZbz76R6XveivnUT9vyuadO=S=+ULRA0=Z0TYWiaY6Wg@mail.gmail.com>
Date: Sat, 30 Jun 2012 14:28:02 -0500
From: Jimmy Hess <mysidia@gmail.com>
To: Todd Underwood <toddunder@gmail.com>
Cc: nanog@nanog.org
Errors-To: nanog-bounces+nanog.discuss=bloom-picayune.mit.edu@nanog.org

On 6/30/12, Todd Underwood <toddunder@gmail.com> wrote:
> This was not a cascading failure.  It was a simple power outage
> Cascading failures involve interdependencies among components.

Actually, you can't really say that.  It's true that it was a simple
power outage for Amazon.
Power failed, causing the AWS service at certain locations to experience issues.

Any of the issues related to services at locations that didn't lose
power are a possible result of cascade.

But as for the other possible outages being reported...  Instagram,
Pinterest, Netflix, Heroku, Woot, Pocket, zoneedit.

Possibly Amazon's power failure caused AWS problems, which resulted in
issues with these services.

Some of these services  may actually have had redundancy in place,
but experience a failure of their service as a result of  unexpected
cascade from the affected site.


> T
--
-JH


home help back first fref pref prev next nref lref last post