[asterisk-users] Teliax Quality of Service

[Douglas Garstang]

> -----Original Message-----
> From: asterisk-users-bounces at lists.digium.com [mailto:asterisk-users-
> bounces at lists.digium.com] On Behalf Of Stephen Bosch
> Sent: Monday, August 06, 2007 9:42 AM
> To: Asterisk Users Mailing List - Non-Commercial Discussion
> Subject: Re: [asterisk-users] Teliax Quality of Service
> 
> Eric "ManxPower" Wieling wrote:
> > Douglas Garstang wrote:
> >> Let's assume for a moment that it's impossible. That does not mean
> adding additional servers and additional networking equipment does not
add
> value, or is a worthless endeavour.
> >
> > I agree with that.  At least two people that I know run ITSPs.  Each
> > time they have an outage (which is not very often) they DO learn
from
> > the experience and work to avoid a future outage cause by the same
> issue.
> >
> > You would be surprised at how many little things can cause an
outage.
> 
> My own experience is that increasing "failover redundancy", which adds
> correspondingly increasing complexity, also increases the odds of an
> outage.
> 
> It is very rare that failover redundancy works as intended during an
> actual failover, no matter how many times you simulate it.
> 
> I would rather have a simple network design where the cause of
failure,
> when it happens, is obvious and quickly corrected. For example, I
would
> rather have replacement parts on the shelf and be able to slap them in
> quickly than be running hot standbys and paying for the electricity,
and
> then have the thing break anyway when there's a failure.

This might work for a web service, but people have a zero tolerance for
no phone service. They expect to be able to pick up their handset, and
get a functional dialtone immediately.

Adding additional servers, additional network components, and some
smarts into your design saves being woken at 3am when a server fails.