Don't go work for that guy! &;-) See more below:

At 03:09 PM 9/14/00, John Barnes wrote:
>I had technical with a CCIE interview yesterday, and

<snip>


>1)      What is the size of a token ring frame?
>My answer: Token ring has a variable frame size.
>His answer: 3 bytes..
>
>Isn’t that the size of the Token frame?

That would be the token.


>2)      What the MTU of a token ring frame?  (Isn’t this
>about the same question as #1?)
>My answer: slightly larger that 16K (I couldn’t
>remember the exact number)
>His answer: about 4470 bytes .

18K on 16-Mbps Token Ring. 4470 bytes on 4-Mbps Token Ring AND FDDI. 
(Technically, both depend on the token-holding timer, which has a default, 
non-configurable value for Token Ring and is configurable for FDDI. But the 
max frame is based on the max the timer could be set to and a network of 
maximum size.)


>Ahh….. what?  He claimed I was thinking about
>FDDI…grrrr  Ah… Who’s thinking about what?
>
>3)      What is the decision making process involved when a
>packet enters a router?  What three criteria are used
>to make this decision?
>        My answer:  It depends. Is this the first
>packet with this destination to arrive at this router?
>  What switching mode is the router configured for.

Excellent answer.


>        His answer:  Forget about that stuff… how does
>it determine which route to use.
>
>        My answer:  longest match in the routing table
>
>        His answer:  What if multiple routes exist in
>the table.
>
>        My answer:  It depends.
>
>        Ok…..I’m gonna cut to the chase… The answer he
>wanted was longest match, Administrative distance,
>then metric.  Ahh…. I’m pretty sure is wrong.   The
>router looks at AD and Metrics long before the packet
>enters the router.  The router uses AD and metric to
>populate the routing table, and then longest match
>from the routing table to make the decision once the
>packet actually enters the router.  Comparing AD and
>metric on every known route every time would place
>unnecessary burden on the CPU.  Compare it once, make
>the decision, and enter it in the RIT.  Even in the
>case of IGRP/EIGRP with variance, the next eligible
>route is determined before the packet enters the
>router.

Well if the router learns a more specific route from a routing protocol, 
that replaces a less specific. To quote Howard, "A summary route from the 
latest, greatest OSPF implementation will be overridden by a RIP subnet 
route from an old UNIX box."

Then AD comes into play. Metrics are used as a tie-breaker for routes 
learned from the same dynamic routing protocol (i.e., breaking ties between 
routes of the same administrative distance).

To move onto frame forwarding, when a route is looked up in the routing 
table, the main criterion is most specific prefix match. For example, the 
most specific possible match is a host route or /32 prefix, while the least 
specific possible match is the default route of 0.0.0.0/0.

See Howard's paper on Routing Principles at www.certificationzone.com for a 
great explanation of how all this works. He's the expert!


>       Maybe I should have picked up on this stuff when
>the recruiter asked me with BGP was a DV or LS based
>routing protocol.  My answer… ahh…neither, it’s path
>vector.

Reminds me of the time years ago when the technical recruiter asked me the 
difference between asynchronous and BISYNC. (He meant to say asynch VS 
synch and had actually never heard of BISYNC. He didn't understand my 
answer and didn't hire me. Just as well, I think!)

Priscilla


>I’m basically sending this out to get thoughts, and
>hopefully Howard, Priscilla or someone can tell me
>wether I’m off technically or not.
>
>
>THANKS!!!!!
>
>-john
>
>
>__________________________________________________
>Do You Yahoo!?
>Yahoo! Mail - Free email you can access from anywhere!
>http://mail.yahoo.com/
>
>**NOTE: New CCNA/CCDA List has been formed. For more information go to
>http://www.groupstudy.com/list/Associates.html
>_________________________________
>UPDATED Posting Guidelines: http://www.groupstudy.com/list/guide.html
>FAQ, list archives, and subscription info: http://www.groupstudy.com
>Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]


________________________

Priscilla Oppenheimer
http://www.priscilla.com

**NOTE: New CCNA/CCDA List has been formed. For more information go to
http://www.groupstudy.com/list/Associates.html
_________________________________
UPDATED Posting Guidelines: http://www.groupstudy.com/list/guide.html
FAQ, list archives, and subscription info: http://www.groupstudy.com
Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]

Reply via email to