RE: Active/Passive "high availability"
- From: Carel-Jan Engel <cjpengel.dbalert@xxxxxxxxx>
- To: Brandon.Allen@xxxxxxxxxxx
- Date: Fri, 13 Apr 2007 23:40:09 +0200
On Fri, 2007-04-13 at 12:29 -0700, Allen, Brandon wrote:
> Summarizing all the below comments as (SAN = SPOF != HA), I would
> argue that it depends on how you define "high availability". The term
> is inherently subjective. Even if you are down 10 days a year, that
> is still relatively "high availability" compared to being down 20 days
> per year. Certainly a fully redundant SAN with a competent admin is
> going to be more highlier (obscure South Park
> reference :-) available than a single spindle with no redundant
> components. A careless, but efficient, tech can bring down multiple
> SANs for quite a long time :-)
>
I've seen 'highly skilled' EMC engineers bringing down EMC DMX boxes for
quite some hours due to 'harmless' upgrades. Cash corruptions, loss of
data, service, whatever you want/don't like. The DMX held just > 100
databases. And they did it twice. In one company. And the first time
they had another case in the same week, the same country, the same
cause.
And of course, Oracle is free of bugs, so one database sitting on
multiple disks or even replicated SANs is fully safe.
Best regards,
Carel-Jan Engel
===
If you think education is expensive, try ignorance. (Derek Bok)
===
Other related posts:
