[AusNOG] AAPT Ethernet outage

[Thomas Jackson]

Usually takes a major outage before many businesses see the value of
redundancy.

 

A good example of this: a site (which hosts a lot of services for other
offices) is fed by a microwave link to a nearby datacentre (no glass in the
ground at the time, and the build cost was going to be too high). This all
worked well until lightning struck the antenna and left a blob of blackened
plastic at the top of the mast. It took us about 12 hours to get it back
online (had to get a new radio, config it, wait for a big enough gap in the
storm to send someone up onto the roof). The total cost of downtime was
astronomical (and, of course, something we'd been warning the business about
for a long time).

 

The next day, we had signed approval to bring in an additional link, set up
BGP, buy new routers to support it etc etc and a directive to "make sure
this never happens again".

 

The probability of an outage is 100% - from backhoe related signal
attenuation to a tech entering the wrong command on the wrong router. The
only difference is how long the outage lasts.

 

 

From: ausnog-bounces at lists.ausnog.net
[mailto:ausnog-bounces at lists.ausnog.net] On Behalf Of Paul Wilkins
Sent: Wednesday, 4 July 2012 10:10 PM
To: ausnog at lists.ausnog.net
Subject: Re: [AusNOG] AAPT Ethernet outage

 

It's not a complicated calculation...

Productivity Value (Expected time to recover) * Probability of outage vs
Cost of providing redundancy

Paul WIlkins

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.ausnog.net/pipermail/ausnog/attachments/20120705/af251f1e/attachment.html>