3 layers cisco design

[jaimedrq1]

Hi friends,

I have a doubt that I hope anyone can resolve.

I am designing a new network following the 3 layers cisco model, and I have the next doubt. (the network is in the image).

Imagine that the root bridge of the Switch Block Usuarios is SW5, and the secondary SW6. Without PVST+, most of the links of SW6 will be blocked.

I am using HSRP to have gateway redundancy in SW5 and SW6, and using in the distribution and core layers RIPv2 to publish the routes.

Imagine that in the Access Layer exists subnets of 172.16.0.0/16, and SW5 and SW6 publish them to SW7 and SW8. So SW7 and SW8 have to ways of same cost to access to this network (1 hop). Thats the problem. SW7 and SW8 will do load balancing, but the SW6 ports are blocked by STP algorithm.

I know that I can use some kind of distribution lists or route maps to solve the problem, but I would like to know how force to use as the best next hop the HSRP router elected like designated.

Thank you very much in advance.

Cheers!

Jaime

[amigomnemonik]

I am not sure what you mean in this statement:

"I know that I can use some kind of distribution lists or route maps to  solve the problem, but I would like to know how force to use as the best  next hop the HSRP router elected like designated.", but IP SLA may be a solution  for the link detection ?

[jaimedrq1]

Hi Kamil, thank you for your answer.

The problem that I have is that core layer have redundant paths to access the 172.16.0.0/16 network (Switch Block Usuarios) leardned from RIPv2.

But the SW5 is the root bridge for this switch block, and SW6 have most of his ports blocked. Then, I don't want that core load charge between this two switches, because SW6 don't switch most of the packets because his ports are blocked.

Then, I would like that the core choose like best next hop to the network 172.16.0.0/16 the switch elected like designated for HSRP like gateway. I have SW5 like HSRP designated switch because this is the root bridge of the network.

Could you tell me how I can to say to the core that choose the HSRP switch designated (SW5) like best next hop to access 172.16.0.0/16 network, or if it's impossible??

Thank you very much.

[Jan Hrnko]

Hi Jaime,

can I kindly ask you which of these links are switched and which ones are routed? Because in such a scenario I would suppose that links between SW5-SW7 and SW6-SW8 (Core network and other links to core switches) should be routed not switched. Thank you very much!

Best regards,

Jan

[amigomnemonik · Accepted solution]

As you are runnning RIP, it uses administrative distance of 120.

What if you put a static route (with administrative distance 1) in the core device and point it to 172.16.0.0/16 subnet and the rediustribute this static across the RIP domain ?

So : ip route 172.16.0.0/16 via SW5

And: ip route 172.16.0.0/16 via SW6 (in case SW5 fails) 2 (2 is a higher administrative distance of the second  static route, the setup of two static routes with different administrative distances is known as floating static route - usefull with BGP to two peers).

The above can also be combined with IP SLA for the ping detection (icmp-replies) for more sensitive detection (if the link does not fail but has a huge packet loss, then most probably the layer two BPDUs will be sent across the SW6)

Regards,

[jaimedrq1]

Thank you very much both Jan and Kamil,

I think that the static routes is a good idea, I don't know why didn't do this before...

I was complicating everything too much... I wanted to mark a route from distribution layer sent from the HSRP router so when the core router receive it put an AD lower than the received from the other switch, but I can accomplish this with manual AD with the static route and everything come easier.

Thank you  very much for your help!!

Have a nice week!

Jaime.