From: James Obot (jamesobot@xxxxxxxxxxx)
Date: Thu May 30 2002 - 08:19:37 GMT-3
CONGRATULATIONS! ! !
J
----Original Message Follows----
From: jsaxe@Crutchfield.com
Reply-To: jsaxe@Crutchfield.com
To: ccielab@groupstudy.com
Subject: CCIE #9376
Date: Wed, 29 May 2002 22:34:34 -0400
(I put a bunch of "bcc"'s on here to various people who I thought might be
interested, but who wouldn't want their email addresses posted forever in a
public forum for spam harvesting. So that's why you're seeing this, if
you're wondering.)
I passed yesterday in Research Triangle -- my second attempt, two-day the
first time and one-day this time. Got the email from their automated system
in the wee hours of this morning. Hurray! I actually felt pretty confident
walking out the door; I mentally flipped through the lab booklet, totaling
up the things I definitely or possibly missed, and I didn't think it could
have been more than 20 points. I guess I was right.
I have to say I like the one-day lab format; it's compressed and
challenging, but they really appear to have succeeded in cutting out the
"fluff" (drawing diagrams, physical cabling) while concentrating on the real
issues. Plus it allows them to schedule more people in the lab, gives
candidates predictable travel plans, and eases the time pressure on the
proctors. I know everyone's going to hate me for saying this, but I believe
it's true: If you absolutely know everything cold, and you get a little
lucky, you can actually finish the whole lab with a half-hour to one hour to
spare. I started down a blind alley once or twice, so I used up my time
margin undoing and reconfiguring -- I just barely finished. If you let
yourself get stuck on one topic, banging your head against it over and over,
the time will evaporate before you know what's happening -- keep pace, and
go on to something else for a while.
Here's a key point, though: You must be at a point where you know with
almost 100% certainty what's going to happen before you hit "return". You
must configure systematically, incrementally, in layers (both OSI model and
the "levels" of Caslow's book), and from one part of your lab network to an
adjacent part. If you just blindly go to three routers of your network,
mutually redistribute 4 IGP protocols, and then sit back and survey the
chaos and try to see what's broken and how to fix it, you'll never find it
all -- recipe for disaster, in my opinion. Just configure a few adjacent
routers, then add one next door through redistribution and make sure it's
stable, fixing problems as you see them. But you should be able to see the
problems coming before they happen; you should see that /28 network over in
the far corner, see that you're trying to hand it off to an IGRP speaker at
the other end of a /24 subnet, and know that someone, somewhere is going to
have to cough up a /24 summary or this guy's never going to hear it.
Remember that routing protocol neighbor formation and normal packet
switching depend on Layer 2 next-hop reachability, and BGP route acceptance
depends on Layer 3 next-hop reachability (among other things). Also remember
(when you're in the exam especially, but even when you're practicing) that
sometimes you're not crazy, and there is a bug in the software, and the poor
little router can't keep up with all the wacky stuff you've done and undone
and redone. Don't be too hesitant to save the config, reload the router, and
go out and get a drink of water or something. Might be when you come back,
it's working just as you expected it to. Reloading costs time, though, so
make sure you have something else to work on while it comes back.
Preparation: The usual books, plus the ECP1 course last year before Mentor
Tech vaporized, and all the playing I could do on my employer's routers
without breaking anything. :-) Also, recently, one week of in-person rack
time at the CCIE Practice Lab at NCSU -- very good, although they didn't
happen to have a crypto image when I went. (I believe they will be working
on this.) And the practice labs in the back of the Cisco Press book "CCIE
Practical Studies, Volume 1"
<http://www.amazon.com/exec/obidos/ASIN/1587200023> are fantastic -- very
much in the spirit of the real lab, designed with the 8-hour time frame of
the one-day format, and comprehensive. I never did purchase any practice
labs, which I slightly regret although of course it doesn't matter now; but
I imagine these 5 labs compare favorably, and it's only a $75 book. Note
that Cisco Press actually has solutions posted on their Web site, which is
also extremely nice, but I found a couple of oversights in the solutions.
(Keeps you on your toes.) They're Adobe Acrobat PDF's, so they could be
corrected.
Thanks very much to my employer, Crutchfield Corporation and Dave Dierolf in
particular, for being incredibly supportive, with time off, training
dollars, and encouragement. It's hard to imagine doing this totally on my
own, always on personal time, buying a rack full of equipment and staying up
late nights typing away at routers all the time. I applaud you dedicated
folks who are doing it this way, who see so much in this goal to pursue it
so diligently. Please remember to put things in perspective, though... go
outside once in a while, and keep in mind that your family is most
important. Thanks so much to my wife Laura, who's probably VERY glad this is
done.
Oh, by the way, IGRP is the devil. Nobody really still uses it in production
networks, right? I'm convinced the only reason it's still implemented in IOS
images is to screw up CCIE lab candidates!
This archive was generated by hypermail 2.1.4 : Thu Jun 13 2002 - 10:59:12 GMT-3