09-01-2004 05:47 PM - edited 03-02-2019 06:11 PM
Does anyone know how routing would work on a 3750 in switch stack mode. I am looking at using two 3750s connected with stackwise running bgp and ospf, routing ospf with F5 BigIP and BGP with 7206.
example of the design:
firewall cluster
| |
3750--3750 stacked with two uplinks to 2 F5s
| |
F5 F5
Thanks,
michael
Solved! Go to Solution.
09-02-2004 02:04 AM
It is the stack master's responsibility to manage all routing functions (control plane). State information is shared between all stack members, however, routing tables are flushed on master role change. I think this is basically done to avoid the possibility of a routing loop. From what i have heard Layer3 failover does not take a considerable amount of time as all the stack members have their routing functionality initialized and the only capability you lose momentarily is learning of new routes. I think the following link will answer most of your questions:
http://www.cisco.com/en/US/products/hw/switches/ps5023/products_white_paper09186a00801b096a.shtml
09-01-2004 09:50 PM
Hi Michael,
when you use two physical 3750 and you use the stackwise technologie, the 3750 works logical as one machine. You can see this with the "show cdp neigh det" command on a neighbor switch. It's also possible to configure a etherchannel from evry chassis to another catalyst. The only restriction is, that you must use etherchannel mode on (PAgP and LAcP doesn't works). The command "show switch" shows you also the information about master and slave switches. And when you have one logical machine you have only one ip routing table.
Regards
Peter
09-01-2004 09:55 PM
Like Peter said the 3750 in a stacked configuration works as one individual device. For all purposes you can consider the stacked 3750 to behave exactly like a single (individual) 3750 would have in the exact same configuration.
09-02-2004 01:52 AM
How does failover work if the master switch is down? Is there shared state with routing on the 3750 when used as a Layer 3 device?
09-02-2004 02:04 AM
It is the stack master's responsibility to manage all routing functions (control plane). State information is shared between all stack members, however, routing tables are flushed on master role change. I think this is basically done to avoid the possibility of a routing loop. From what i have heard Layer3 failover does not take a considerable amount of time as all the stack members have their routing functionality initialized and the only capability you lose momentarily is learning of new routes. I think the following link will answer most of your questions:
http://www.cisco.com/en/US/products/hw/switches/ps5023/products_white_paper09186a00801b096a.shtml
Discover and save your favorite ideas. Come back to expert answers, step-by-step guides, recent topics, and more.
New here? Get started with these tips. How to use Community New member guide