From: Jerry Haverkos (jhaverkos@xxxxxxxxxxxxxxx)
Date: Thu Jul 04 2002 - 10:59:22 GMT-3
Point(s) taken. I appreciate your input.......jerry
-----Original Message-----
From: Jake [mailto:jakeczyz@yahoo.com]
Sent: Wednesday, July 03, 2002 8:41 PM
To: Jerry Haverkos
Cc: ccielab@groupstudy.com
Subject: RE: OSPF Demand Circuit/PPP Multilink/Dialer Profile
Hi Jerry,
I'm sure someone else will answer with more detail, but both actually
provide great
flexibility in configuring multiple dial legs. Also, if you configured the
physical
interface instead of the logical one, you would be limited in your ability
to utilize
that interface for other connections. But, depending on your particular
need, the
flexibility is there. If you simply need a dedicated point-to-point backup,
then maybe
you wouldn't use it. This is by no means an attempt at an extensive
justification, just
prima facie.
--- The following is a paid advertisement by sysco: ----
As for the exam itself, (please note that my tone here is playful and not
argumentative),
would you also argue that high schools shouldn't teach trigonometry or
"college" algebra
or even chemistry, since probably less than 1/10 of 1 percent of people
actually use it
in "real life?" Or, perhaps the SAT's verbal section should only test
people's mastery of
the most common words (not "prima facie," for example)? Obviously, the
challenge lies at
the edges, the edges of the IOS and, for that matter, the "edge routers" are
where the
biggest challenges lie.
So, to be fair, in the "real world" you would not be running 5 different
routing
protocols and maybe even a sprinkle of IRB, with a touch of voice - all on 6
routers.
But, if you had to do these complex tasks and make them work quickly (like
when my
company lost $5 MM in 8 hours because a Sup went down which would have cost
$30 K to make
redundant... but I digress)... you would want someone that knows how to push
the IOS to
its limits and make it do what's normally not possible/practical/logical,
etc. I hate to
burst any bubbles, but after you've mastered all that, the "day-to-day" of
running a
network is relatively easy.
Without trying to sound arrogant or condescending, here's an analogy I used
to describe
it to a buddy at work: Imagine you're job is to paint a box. Painting the
edges is the
tough part, while the flat surface is relatively easy. Now picture a box the
size of a
football field. Relatively speaking, there is a whole lot more flat than
there is edges.
So, you don't learn too much about painting edges after years on this job.
The exam is
like a tiny box, which is mostly edges. (In effect, you may have 6
redistributing "edge"
routers) So, by training, practicing, and proving yourself on the tiny box,
your ability
to tackle the "real world" boxes of most corporations, which are huge - is
solid. This
isn't a flawless analogy, but it's the best my tired brain could think of.
I know you're not one of these, because you mentioned that you "heard" it
was not "real
world," but I've found that most of the people who complain about the test's
relevance
are people with years of experience "painting the middle" and wondering why
they can't
pass the exam after a couple attempts. I could be wrong.
(Four out of five CCIE's agree, your mileage may vary. Sense of humor,
required)
Keep the faith,
Jake
9102
--- Jerry Haverkos <jhaverkos@columbus.rr.com> wrote:
> I've heard the CCIE lab doesn't necessarily present real world situations.
> Why would you want a demand circuit defined with a dialer interface? It is
> basically a dedicated backup circuit isn't it?
>
> On the "Corporate/Host" side, I don't believe you'd want to define a
dialer
> profile either. Comments?
This archive was generated by hypermail 2.1.4 : Sat Sep 07 2002 - 19:36:18 GMT-3