cancel
Showing results for 
Search instead for 
Did you mean: 
cancel
544
Views
5
Helpful
7
Replies

Dual Core Redundancy.

Hi ,

We are trying to setup a Dual core redundant network. We currently have two 8540's forming the core layer and 6509;s forming the distribution layer. What we are trying to is to have a fully redundant core.Our current design has the two coresss connected to the 6509's using port channels.This is a one to one connection .. ie CORE1 to 65091 and CORE2 -- 65092. What we want to do is to have connections from CORE1-- 65092 and CORE2-- 65091 while still maintaining the existing connections. We are running HSRP and have the vlans split up between the two cores. VLANS 2-8 are local to core1 while VLAN's 9-17 are local to Core2 . While we know that there is a way to do this ... the problem comes up if two diagonal devices fail ... say CORE1 and 65092 . What will happen to the vlans on CORE 2 ?

I would appreciate any design suggestions on this problem. I will try to paste a diagram explaining the situation .

Thanks.

Karthik.

7 Replies 7

ccobtn
Level 1
Level 1

Hi,

First is the traffic purely layer two or Layer 3. If it is layer 3 what you could do is created all you vlans on both core Switches and than use HSRP between the vlans i.e. Core 1 is active for Vlan 2-8 and Backup for Vlan9-17, similarly Core2 is active for Vlans 9-17 and backup for 2-8. In this way if one of the Core 1 and 65091 fail, traffic will be re-route via core2/65092. For this to work I assume that devices behind Core Switches are dual Ethernet, also same applies to you access switches???.

That is the way we have it setup right now. I know that if CORE1 and 65091 fail then the traffic would be redirected through CORE2 and 65092. The trouble is what happens when CORE 1 and 65092 fail or if CORE2 and 65091 fails.

Hi,

In this case what you could do is that D1 will have link to both Core Switches as well as D2, also D2 will have link to both Core Switches. Di will have Core1 as a Master and Core2 as a Backup via static routes/routing protocol/hsrp, and reverse for D2. Now if Core1 and D1 fail, D2 will forward all the trafics to Core2. Now you have full Redundancy

Regards

I agree... but my question was what would happen say for instance in the event CORE 1 and D2 fails ... instead of CORE 1 and D1.

Sorry what I meant was that if D1 is linked to both Core Switches as well as D2, same for the D2. Now for D1, if the link to Core1 and D2 fails traffic will be re-routed via Core2. This gives you high a availability as long as there is two devices available i.e. Core1/D2 fail, D1 will re-router traffic via Core2, Core2/D1 link fails D2 will re-route traffic via Core 1. The only exception is if both Cores and one of the Dest fail.

Hope this helps.

I do not think there would be a problem if you had a diagonal Switch failure (say Core1 and 65092), Core2 would know to send traffic for the odd Vlans to 65091 cause of HSRP.

Thanks for your help guys ... I'll try some things out this weekend and report .

Getting Started

Find answers to your questions by entering keywords or phrases in the Search bar above. New here? Use these resources to familiarize yourself with the community: