interview question...

sivakondalarao · ‎04-02-2006

hi,

in interview i have faced this question.

question: I have 200 servers they are connected to 20 layer2 access switches (each server connected to two switches for redundancy)

20 layer2 access switches are connected to 2 layer3 distribution switches. (each access switch connected to two distribution switches)

2 layer3 switches are connected to two routers, where each router is connected to one ISP.( each distribution switch connected to two routers)

If I want no single point of failure in the network, what are the protocols to be used on each device?

on servers?

on access switches?

on distribution switches?

on edge routers?

Regards

skrao

mahmoodmkl · ‎04-02-2006

Hi

what sort of redundency they want for example for gateway redundency u can go for HSRP at u r distribution layer if u want.

and on access switches u can go for UPLINKFAST.

on servers i think there is no need.may some others have different view.

and on edge routers u can go for VVRP or GLBP

Hope this helps

Mahmood

mheusinger · ‎04-02-2006

Hello,

I would suggest:

servers: NIC teaming (active/standby for the server IP!)

between servers and L3 switches: HSRP/VRRP/GLBP (redundant default gateways)

between L2 and L3 switches: Rapid STP (needed because of NIC teaming and L2 redundancy)

between L3 switches and edge routers: OSPF (could be any dynamic IP routing protocol able to handle the default route to the internet)

between edge router and ISP: eBGP

between edge routers: iBGP

Hope this helps! Please rate all posts.

Regards, Martin