From: Pamela Forsyth (pforsyth@xxxxxxxxx)
Date: Fri Aug 13 1999 - 11:41:41 GMT-3
Howard,
Not witchcraft. Total possible number of points for the two days is 100.
The number of points to pass is 80. The points allocated to each day's
tasks, and the number of points for a given task, vary with the exam
version. The proctor will tell you how many points are required to get
through each day before you start the lab, and each task and subtask is
labelled with the number of points you will accrue for completing it so you
can keep track and determine your strategy.
Some tasks, obviously, MUST be completed for the rest of the lab to work.
These are usually dealing with getting the physical interfaces to come up,
and with proper configuration of data-link layer technologies and
appropriate network-layer addressing. But you have complete discretion in
determining the order in which you complete the day's tasks.
There are two critical points at which your work is graded--one at the end
of the first day, and the second halfway through the second day, just
before troubleshooting. In order to proceed to the second day or to the
troubleshooting, you must have scored at least enough points so that if you
manage to score all possible points for the remaining tasks, you could
still conceivably pass. This means that you could run a 20-point deficit
either at the end of the first day, OR going into troubleshooting. But
realistically, if you don't have any spare points going into the second day
you don't have much of a chance of completing the lab successfully. My
experience was that the second day was much more difficult, not so much
technically but in terms of time management.
If you are at a 20-point deficit going into troubleshooting you do have a
prayer, but the pressure will really be on. My proctor didn't tell me how
many problems he had put into my network, so I really couldn't be certain I
had found all of them.
I didn't concern myself too much with keeping track of points when I did
the lab--I don't think it's productive to worry about it. I recommend you
not waste any time adding up points--better to spend that time ensuring
your configs are working the way you want them to, and **documenting** your
network. The documentation is worth almost nothing in terms of points, but
you won't be able to pass the lab unless it is done accurately--unless you
are one of those rare individuals with photographic memory.
The procedures have changed a lot over the history of the program. It has
been getting progressively tougher over the years. One very early CCIE
told me they were able to pair up in teams. As recently as May of 1997,
the RTP lab was still providing both days' tasks the morning of the first
day, giving candidates an opportunity to review that evening any
protocols/technologies they were able to put off until day 2; there was
also "all or none" scoring of tasks. Each task would have several
subtasks, but you wouldn't get any points for the task unless you completed
all subtasks successfully. Now it is possible to score the points for the
individual subtasks you are able to complete, even if you can't complete
all the subtasks under a task heading.
They stopped letting people bring their personal notes into the lab around
September or October 1996, AFAIK. The evolution of procedures and
restrictions may account for differing stories you may have heard.
I was encouraged to stay and work with the equipment and the lab problems
after failing the lab--you should ask for this privilege if it is not
offered. The rack has been allocated for your use for the full two days,
and it is good to take advantage of it. The proctor will not give you any
hints about the tasks you were not able to complete, nor should you expect
him to.
My successful lab experience was in February 1998. The exam content has
changed, but I don't think the procedures have changed very much since then.
Pamela
At 10:01 PM 8/12/99 -0400, Rahmlow, Howard F. wrote:
>Anybody out here know how the CCIE lab is scored. I have seen all the scary
>options, but not 2 match. I know you need 80 to pass, that seems to be about
>it. The rest is witchcraft.
>
>Thanks, Howard.
>
This archive was generated by hypermail 2.1.4 : Thu Jun 13 2002 - 08:21:46 GMT-3