CCIE Lab Preparation

Just another CCIE blog

  • Category

  • Archives

  • Advertisements

IEWBv4 Lab 4 – Core Tasks

Posted by Jo on October 17, 2007

I had an impromptu 4 hour session this morning with Lab 4. My new aim is to get through as much as the core sections as possible. This usually covers the following:

Bridging and Switching
Frame Relay

I made a great start with the troubleshooting, and spotted 3 of the 4 deliberate mistakes in the initial configs. I cracked through setting up the trunk links between all of the switches and creating VLANs on the STP Server. There were no special requirements listed in the task for anything that required client or transparent mode within the VTP domain.

There were a couple of tasks based around manipulating the root bridge for each VLAN, and setting costs on interfaces that I had to check up on, but know I know the best way to determine these – hopefully. Nothing else in this section caught me out too much, and was easy enough to configure with the ? command or by the DocCD.

The Frame-Relay section had a couple of interesting requirements – basically a hub and spoke design between R1, R2 and R3 with R2 as the hub. R2 has a physical interface , and no sub-interfaces were allowed to be created on R2. Alternatively R1 and R3 were not allowed static (or dynamic) layer 3 to layer mapping allowed on the hubs. As no dynamic or static mapping is allowed on R1 and R3, these had to be configured as point-to-point sub interfaces with the frame-relay interface-dlci xxx command configured to enable communication across the frame relay cloud. R2 was configured normally with frame-relay map ip commands for each hub with the broadcast keyword.

There was a one point question about configuring a PPP serial link to maximise efficiency by guessing character streams that are being sent over it.I worked this one out, by typing compress ? and was presented with the predictor keyword, so I was pretty sure this was the answer.

Onto IGP and OSPF. Due to the physical layer 2 topology there were a few interesting tasks to work through for OSPF. I had to run OSPF over R1, R2 and R3 – remember R2 is the hub – and wasnt allowed to use the ip ospf network interface level command on R2. This meant I could only change it on R1 and R3. A quick check of the interface type on R2 with show ip ospf interface showed me that it was defaulting to NON_BROADCAST so I had to match the interfaces on R1 and R3 with this. Another caveat with this network type is that there is a DR/BDR election, so I had to make sure that the hub routers never tried to become the DR/BDR. This was accomplished by setting the OSPF priority on each interface to 0 as follows: ip ospf priority 0.

There most of the other tasks were fairly straightforward, the usual advertise a loopback interface into OSPF but dont use the network statement. This is usually done with a redistribute connected subnets route-map XXXXX where the route-map matches an interface or interfaces. I also had to set up one of the area’s so that it didn’t see any inter-area or external routes. This needed to be a Not So Totally Stubby area (as I was redistributing a loopback into it already) so I configured the area x nssa no-summary on the ABR’s and also remembered to configure area x nssa on the routers internal to this area.

The next task was my first encounter setting up a tunnel interface. The task required that area 2 be configured on R4 and R5, the problem was that R5 only in area 1 (and now area 2) so there was no link back to area 0 as OSPF requires. The question for the task stated an additional subnet could be used to accomplish the task. I had to configure interface Tunnel0 on R5 with a source of F0/1 and a destination of the ABR that connects to area 0 (in this case R2). The reverse config was applied on R2 for the Tunnel interface, and once I had added the tunnel interfaces into OSPF area 0 on each router the routes showed up in the table. On reading through the solutions guide, It stated that usually a virtual-link can be used to join an area to area 0, but in this case it would have needed to cross a stub area (my Not So Totally Stubby from above) so when that happens a tunnel is used instead – pretty cool stuff to know.

The IGP redistribution task was OK this time. No multiple redistribution points to worry about, so no need to tag routes or play around with changing distances etc. This one revolved around how routes were seen throughout the OSPF domain. Firstly I had to create a summary-address for some routes learned by RIP from BB3 – these subnets needed to be visible across the OSPF with a cumulative metric.I also had to redistribute a subnet (VLAN 43 in this case) into OSPF and make sure that all devices saw it with a metric of 100.

This was achieved by creating a prefix-list to match the VLAN43 subnet

ip prefix-list VLAN43 seq 5 permit

Then the route-map was constructed as follows:

route-map RIP-OSPF permit 10
match ip address prefix-list VLAN43
set metric 100

route-map RIP-OSPF permit 20
set metric-type type-1

This was then applied to the redistribute command within the OSPF process

redistribute rip subnets route-map RIP-OSPF

The effect of the route-map is the first entry sets the metric to 100 for the VLAN43 subnet when it is redistributed. By default external routes are type-2 (E2) when going into OSPF, which means the metric remains the same through out the OSPF domain.

The permit 20 entry sets all other routes that are being distributed into OSPF to have a metric-type of type-1 (E1) which means the metric will change as it goes through the OSPF domain.

I moved onto BGP and had enough time left to set up all of the neighbour relationships. There were no Route Reflector or confederation requirements so this was quite straightforward.

I then ran out of time, but all in all I was happy with the way I approached the tasks. I was happy that I saw a couple of new concepts in todays lab (Tunnel interfaces over stub areas, and OSPF network types). I have got my configs saved, so I will finish of the BGP section next time (2 questions to go) and then move onto Lab 5.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: