I have a pair of 11503 units configured with Box to Box redundacy which have been working well for the last three years using box to box redundancy.
We are now commissioning a new disaster recovery data centre which is a few miles away from the primary site. I have the facility to propogate vlans between these sites and was hoping to simply take the standby 11503 to the remote site and extenend the production and redundancy vlans to that site.
In the disaster recovery design we have, Box to Box would work ok so Its an easy win for me.
However, the config guide recommends that the redundancy vlans are connected with a cross over cable back to back and not to use a layer 2 device in the path. Clearly in my situation, this would involve the redundancy vlan being propogated over at least two switches.
Question is this , What is the issue with running this over a switch ?, presumably something to do with mac addressing, but what exactly ?
And am I right in thinking that I am going to have to reconfigure redundancy on these units from box to box, to VIP/VRRP ?
there is no *issue*.
It's just that the protocol was designed to work with CSS directly connected to each other and therefore we only support this type of deployment.
I believe some people use the L2 switches without any problem.
Also, personally I always recommend vip/interface redundancy over box-to-box because you get faster failover and L2 switches are ok between the 2 sites.