Layer 3 network design

nelloconnellocon Member Posts: 25 ■□□□□□□□□□
Hello there,
I hope I am asking this question in the right place...

Anyway I am designing a new network for my first time and I will be designing it as attached using the Layer 3 model. However, i have a few questions that I hope you can help me with as I cant quite fully understand it....

Question 1: What is the reasoning in using a point to point network between the routers and the switches. Why not just use EIGRP?

Question 2: How will the switches know which router to use primarily. One of the routers is the primary circuit and the other is backup? How will it know to failover if the primary goes down?

Sorry for the basic questions but I just need to get some things right in my head.

Thanks all.

Comments

  • malcyboodmalcybood Member Posts: 900 ■■■□□□□□□□
    nellocon wrote: »
    Hello there,
    I hope I am asking this question in the right place...

    Anyway I am designing a new network for my first time and I will be designing it as attached using the Layer 3 model. However, i have a few questions that I hope you can help me with as I cant quite fully understand it....

    Question 1: What is the reasoning in using a point to point network between the routers and the switches. Why not just use EIGRP?

    Question 2: How will the switches know which router to use primarily. One of the routers is the primary circuit and the other is backup? How will it know to failover if the primary goes down?

    Sorry for the basic questions but I just need to get some things right in my head.

    Thanks all.

    Hi,

    Think you need to do some reading my friend.

    Some basic topics to read up on based on your diagram:

    - HSRP (redundancy and router selection outbound)
    - Etherchannel (for the two links between switches)
    - OSPF, EIGRP, iBGP (for the routing protocol and path selection)

    The answers to your questions are:

    Question 1: What is the reasoning in using a point to point network between the routers and the switches. Why not just use EIGRP?

    Point to point links are configured with point to point IP addressing i.e. 10.0.0.0 /30 where you would assign 10.0.0.1 to one side of the link then 10.0.0.2 to the other side of the link.

    In order for the rest of the network to know these exist you need to either configure static routing which I wouldn't recommend or run a routing protocol such as EIGRP or OSPF then advertise the network.

    Do you have any routing or networking experience and do you have anyone to help you with this?

    Question 2: How will the switches know which router to use primarily. One of the routers is the primary circuit and the other is backup? How will it know to failover if the primary goes down?

    There are several ways to do this but you would use a combination of HSRP and a routing protocol configuration to force traffic down a specific link. I had a fairly generic HSRP diagram in my "archive" which explains at a high level how HSRP operates and also attached a point to point core to distribution link diagram similar to what you have, except the core is the 2 middle switches and distribution are the 6506's and NEXUS switches in the diagram.

    OSPF or EIGRP would be enabled for these point to point links and any user networks that would connect to the distribution layer (access layer).

    If all of the above sounds like mumbo jumbo I think you need someone to help you design this.

    Hope it helps
    Malc
  • nelloconnellocon Member Posts: 25 ■□□□□□□□□□
    Hi Malcybood,
    Thanks for taking the time to reply. I actually do understand a little HRSP. What I have struggled to understand and can never get someone to quite explain it to me has to do with the interface vlan (not sure where or what to read up on for this)

    The VIP address .1 exists on the layer 3 switches. So lets say int vlan 1 has an ip of 10.0.0.2 on one switch and 10.0.0.3 on the other switch and the VIP is .1. How does traffic hitting the VIP address know which router to choose?

    I can understand where the two routers are .2 and .3 and the VIP is .1 for either of them but what about when the VIP is on the switches.

    Am I making any sense...im even frustrating myself icon_sad.gif
  • pham0329pham0329 Member Posts: 556
    nellocon wrote: »
    The VIP address .1 exists on the layer 3 switches. So lets say int vlan 1 has an ip of 10.0.0.2 on one switch and 10.0.0.3 on the other switch and the VIP is .1. How does traffic hitting the VIP address know which router to choose?

    I can understand where the two routers are .2 and .3 and the VIP is .1 for either of them but what about when the VIP is on the switches.

    Am I making any sense...im even frustrating myself icon_sad.gif

    Need to read up on HSRP some more! The active router in the standby/hsrp group handles all incoming traffic forwarded to the VIP/virtual mac address.

    So, in your example, suppose a device is sending traffic to a remote network
    1. The device compares the destination IP to its own local IP and knows that it needs to send the traffic to the default gateway
    2. The device sends an ARP request to 10.0.0.1 (its GW)
    3. The active router sees the ARP request, and respond back with the virtual mac address 0000.0c07.acxx (i think..), where xx is the hsrp group #
    4. The host sends traffic to that mac address, and is processed by the active router in the hsrp group.
  • ZartanasaurusZartanasaurus Member Posts: 2,008 ■■■■■■■■■□
    Who are you designing a network for?
    Currently reading:
    IPSec VPN Design 44%
    Mastering VMWare vSphere 5​ 42.8%
Sign In or Register to comment.