Hi, I have:
- 1 Load balancer
- 1 Pool on the load balancer containing two IIS servers.
- 2 IIS servers
- 1 SQL Server
- 1 File server
- two 3550 Catalyst switches.
- Several VLANs configured the same on both switches.
- the two IIS servers are dual-homed to the frontend and backend. The frontend is on the loadbalancer's network. The backend is on the corporate network.
- The sql and the file server is homed only to the backend.
- The backend nics on each server is first in the binding order.
- No default gateway is configured on the backend nics.
- Default gateway on the webservers are on the frontend nic and point to the loadbalancer.
- Persistent routes are configured pointing to corporate network.
Forgive me if all of this sounds basic. I know I should have two load balancers, but I only have one. I'm trying to configure a high availability solution for the web application which consists of 2 loadbalanced webservers, a sql server, and a file server as listed above, such that if one of the switches fail, traffic would be routed through the other switch.
But I'm not sure of the best way to do this. I have the same VLANs configured on both 3550's, and I feel I should cross-connect the VLANs, but I'm not sure. If so, how do I plug the servers into them. One web server on the one switch, and the other web server on the other switch? That sounds good, but I only have one load balancer. What if the switch which the load balancer is plugged into fails? See my dilemna? Or do I dual-home the servers to each switch? If so, that would give them two IPs, wouldn't it? which won't be good for name resolution. So as you can see, I need some help. I'd appreciate it if someone can give me advice. Thank you kindly in advance.
