Redundant Network Setup

[sixarm]

Hi,

We've recently taken delivery of a c7000 BladeSystem Enclosure and I'm in the process of getting it up and running and integating it with our existing VMware vSphere environment.

At present all I have installed are the 2 OAs, 1 BL460c G7 and 2 GbE2c Layer 2/3 Ethernet Blade Switches in Interconnect Modules 1 & 2. I've connected Port 20 on each switch into our existing network each one to a seperate switch.

I've installed ESX 4.1 U2 on the blade local HDD and this is up and running. I'm able to see the two NICs within vSphere. However I'm having a few issues with redundancy.

I kept losing ping to one of the Blade Switches (Switch1), however pings to the ESX host were still successful. I had to reset the switch from the OA to regain management and connectivity and in doing so I lose connectivity to the ESX host. What I would expect to happen is for traffic to route via Switch2. I've also tested this by disabling Port 1 on Switch1, and again I lost connectivity to the host completely.

If I disable Port 1 on Switch2 I remain connected to the host and the disabled port is reflected with vSphere by a red cross against vmnic1.

The switches are pretty much default configuration with no VLANs in play.

Any ideas?