From: Carlos Trujillo (carlos.trujillo.jimenez@gmail.com)
Date: Tue Apr 29 2008 - 02:08:38 ART
Thanks Patrick for the answer, but what about the lacp option? The task
states that this trunk must allow lacp negotiation. so where do I apply the
lacp negotiation if I create two separate trunk ports?
I think, this is a tricky one! :)
2008/4/29, Patrick Galligan <pgalligan@gmail.com>:
>
> You need to create 2 separate trunks fro f0/16 & f0/17, via dot1q
> tunnels. Then modify spanning tree to meet the other requirements.
> These questions are a good example of why you need to read the whole
> exam before configuring things :)
>
> On Tue, Apr 29, 2008 at 2:44 PM, Carlos Trujillo
> <carlos.trujillo.jimenez@gmail.com> wrote:
> > Hi Group!
> >
> > Please if someone here can help me clarifying or understanding what
> really
> > is asked to do in the following two questions!. Im pasting exactly the
> > content of both questions directly from the workbook Im using.
> >
> >
> > The way the questions are made, is not so clear for me, or perhaps I'm
> > misunderstanding something here. What Im trying to said, is that in the
> > first question I understand what to do, and I choose a solution to
> > implement, but after reading the second question, I doubt about the
> answer I
> > choosed in the first question,Im not so sure if the answer I choosed
> from
> > the first question is the correct or not! ...
> >
> >
> > [SW1] fa0/16------------------fa0/13[SW3]fas0/16----------fa0/16[SW2]
> >
> fa0/17------------------fa0/14....... fas0/17-----------fa0/17
> >
> >
> > 1.7. DOT1q TUNNEL
> > Configure a trunk between interface Fa0/16 & Fa0/17 on SW1 and SW2 as
> > 802.1q trunk links. This trunk must allow LACP negotiation.
> > SW1 & SW2 must be able to run CDP through this trunk and each other in
> > CDP.
> >
> >
> >
> > **** POSIBLE SOLUTION:
> >
> > the first solution that comes to my head is to create a dot1q
> etherchannel
> > between SW1 AND SW2 using lacp protocol at both sides, the
> etherchannel
> > muts be formed using fastethernet 0/16 and 0/17 ports. Now for the
> > etherchanneling to work from sw1 to sw2, the ports in sw3 must be set
> for
> > dot1q tunneling, and tunneling cdp and lacp.
> >
> >
> >
> > 1.9. SPANNING-TREE PROTOCOL
> >
> > All traffic between SW1 and SW2 for these VLANs should transit the
> trunk
> > between SW1 and SW2's port Fa0/16.
> > In the case that port Fa0/16 goes down, traffic for these VLANs should
> > transit port Fa0/17.
> > As a last resort traffic for these VLANs should transit port Fa0/13 on
> SW1
> > and SW2 if both of the other trunk links are down.
> > This configuration should be done on SW1.
> >
> >
> >
> > **** POSIBLE SOLUTION MAKING DOUBT THE FIRS SOLUTION USED EARLY:
> > After carefully reading the second question, It makes me understand
> that an
> > etherchannel was not asked to do in the first question as per the
> > requirement states: "All traffic between SW1 and SW2 for these VLANs
> should
> > transit the trunk between SW1 and SW2's port Fa0/16,
> > In the case that port Fa0/16 goes down, traffic for these VLANs should
> > transit port Fa0/17"
> >
> > After reading the task 1.9 I can see it is contradictory to my
> > first solution I choosed in the task 1.7, because If I choosed
> > etherchanneling, an l2tunneling lacp frames the traffic will flow in
> both
> > interfaces : fast0/16 and fas 0/17, and it is violating the requirement
> of
> > task 1.9 who ask to transmit traffic only via fast0/16.
> >
> > So, as a final solution to try to satisfy all what is answered without
> > violating any requirement, I think, that I must create an
> etherchanneling of
> > fastethernet 0/16, with a group different than the etherchanneling of
> > fastethernet 0/17 at both switches SW1 AND SW2. and in SW3 to l2 tunnel
> the
> > lacp right??
> >
> > Please if someone agrees with my last result or has any other ideas,
> you are
> > welcome!
> >
> >
> > Pass the CCIE in six weeks, Guaranteed!
> > http://www.certscience.com/CCIE
> > _______________________________________________________________________
> > Subscription information may be found at:
> > http://www.groupstudy.com/list/CCIELab.html
> >
> >
> >
> >
>
>
>
> --
> Certain techniques work on a motorcycle. Other techniques don't, and they
> hurt
Pass the CCIE in six weeks, Guaranteed!
http://www.certscience.com/CCIE
This archive was generated by hypermail 2.1.4 : Thu May 01 2008 - 08:25:52 ART