From: Charles Henson (chenson@gmail.com)
Date: Mon Oct 20 2008 - 23:22:37 ARST
Wow. That's good stuff. Thank you and we see who all replies with more
insight. Thanks.
On 10/20/08, Michael Dorion <dorionm@gmail.com> wrote:
> I've been twice now, and I am also taking the exam again next month.
> Everyone says to 'draw your own diagram and read the entire exam,' but i
> think you should just kind of have an idea what order you should do things
> in, and know what can cause you to have to backtrack. Skim the entire lab
> for these kinds of things. I was actually just starting to write about this
> in my blog.
>
> When I do practice labs I follow these steps:
>
> 1.
> Draw a 'spreadsheat' with all task numbers and points. leave a comment
> section next to each task.
> 2.
> Make a copy of the initial config. 'copy run flash:/initial.config' and look
> it over for any obvious troubleshooting problems like wrong IPs, frame relay
> mappings, etc. If you do run into problems you can use the "more" command on
> the file to see if there is anything that jumps out at you. When I am doing
> this, I add all of my 'default' config on each device, here is what I
> typically paste in to each device.
>
> conf t
> do clock set 20:28:30 20 Oct 2008
> logging con debu
> do debu ip routing
> ip routing
> no ip domain-look
> ip rcmd rsh-en
> ip rcmd remote-host RackxxR1 <#.#.#.#> RackxxR1 enable
> ! this is for TCL scripting later and doesnt need to be on R1
> logging buff debu
> alias exec ir sho ip route
> alias exec ib sho ip int bri | ex unas
> alias exec r sho run | be router
> do wr
>
> 3.
> Start skimming the lab. Pay VERY close attention to the 'rules of
> engagement.' This will tell you what solutions you have open to you and even
> give you hints as to what problems you might run into. I make comments on
> the spreadsheet for problems I see, or rules I should be careful of breaking
> as I skim. Also, I dont draw a layer 3 topology diagram unless I need to,
> but I ALWAYS draw a layer 2 diagram that shows all trunks, VLANs, etc.
> Drawing a layer 3 diagram usually ends up being a huge time waster. Just get
> used to using other peoples' diagrams.
>
> Once i've done that I start approaching the switching section as a whole and
> read the entire section for switching. There isn't much in switching that
> can mess up your routing and other stuff as long as it works (MTU sizes,
> port security, etc), but there are a ton of ways to order the tasks WITHIN
> switching to mess you up. Just look at the internetwork expert workbooks,
> lab 8 for example. As a rule, you do you trunking and vlan assignments
> before you bother with VTP.
>
> Here are some things I look for in a skim:
> -If you see a dot1q-tunnel task when you are skimming, change the MTU to
> 1504 right there and reload. That way its done by the time you are ready to
> work on configs. (same for frame relay mappings to 0.0.0.0).
> -Look for your partial-mesh frame-relay. I dont think i've seen many
> practice labs without them, you should know by now all the problems you'll
> face
> -Split horizon?
> -Discontiguous area 0's for OSPF. (can i add IPs to do a dot1q tunnel? do I
> have authenticat my Virtual Link? is unnumbered a possible solution?)
> -BGP problem? next hop issues, loopback peerings, TTL issues for exernal
> peerings.
> -RPF failures if multiple paths?
> -Problems with ACLs or other security features breaking other configs? bgp,
> gre, ip services etc. I usually read the security section again at the end,
> I've been burned too many times by this.
> -Anything you don't know, and isn't core, mark it to be skipped. You want to
> have a good 60 points by lunch time. (It will taste much better). :)
>
> Hope this helps. I look forward to reading other peoples suggestions.
>
-- CharlesBlogs and organic groups at http://www.ccie.net
This archive was generated by hypermail 2.1.4 : Sat Nov 01 2008 - 15:35:21 ARST