From: CCIE (ccie@madisonsolutions.net)
Date: Sun Dec 17 2006 - 11:02:39 ART
Good morning:
I concur, change is disruptive for a short period. Playing with four
switches in my own lab I have started to see things I did not expect and
wanted to pass along.
We all have read about VTP and when you add a new switch to a network you
should set the VTP to transitive before bringing the switch on line. The
other day as I added a switch into my lab to practice spanning-tree, and the
VTP election of the additional switch deleated all of my VLAN's.
Having never seen this in a real network I was taken back for a moment. All
of the interfaces with assigned VLAN's went from happy blinking green to
flatline orange. My network went from working to dead.. All of the VLAN's
that I had built were still on the interfaces but not in the vlan database.
I discoverd the issue after a moment and resolved the VTP election by having
only one "vtp mode server" in the switch domain. I mention this as an
example of what can go wrong on the new lab, or worse what is designed into
the new lab. With both 3550's and 3560's it would be very easy for the
network to go dark in the middle of your lab because of a VTP issue.
For the Lab I think it is wise to not only look at configuration issues, but
to remember the simple rules to stop spanning-tree loops, check what you are
doing as you bring up interfaces, vtp issues, etc. The big clue that I
expect to see on the Lab, is that the trunking interfaces between switches
are shutdown.
Any port that is dark that needs be turned up is going to be suppect, and
should be analized before turning up. The second clue would be to watch to
see the VTP commands, and the order in which you activate them. If the
VLAN's are pre-built and the primary VTP server will become a switch with no
VLAN's; what have you added to your lab panic time?
The last tip before I go to breakfast, study the switch topology before
starting to configure the swith network. Expect all sorts of loops,
mis-configured interfaces, etc.
Merry Christmas and be loop free in 2007.
George Morton, Ph. D.
----- Original Message -----
From: "Darby Weaver" <darbyweaver@yahoo.com>
To: "'Cisco certification'" <ccielab@groupstudy.com>
Sent: Saturday, December 16, 2006 12:43 PM
Subject: Prediction - Lab Results
>I know that there have only been a very few people who
> have passed the lab since the 3-4 switches were
> introduced into the lab this last month or so...
>
> It seems it is always this way after a change, no
> matter how minor.
>
> However, I want to be the first to say that I predict
> that Lab Candidates will begin to pass the lab again
> at or near the same passing rates as before (the new
> labs were introduced last month)as early say...
> within 30 days or so.
>
> I just have a gut feeling... about this.
>
> Anyone care to watch the numbers with me...
>
> Funny how it is all a numbers game in the end.
>
> Merry Christmas and Happy New Year!
>
>
> And I can't wait till my favorite vendors update their
> materials on Switching as well. Not that what they
> have is not pretty good already, but I'm sure they are
> not sleeping at the wheel either.
>
>
> Meanwhile, has anyone taken a look a Cisco Press
> Practical Studies for Switching.
>
> It offers a lot of great insight if you can look past
> the model numbers and get the same great results from
> some 3550/3560's.
>
> _______________________________________________________________________
> Subscription information may be found at:
> http://www.groupstudy.com/list/CCIELab.html
This archive was generated by hypermail 2.1.4 : Tue Jan 02 2007 - 07:50:38 ART