Share your SPOF experiences using blade servers

[BenConrad]

All,

I'm interested in finding out if anybody has implemented a blade solution and then subsequently experienced an Single Point of Failure (SPOF) issue.

I'm interested in the sort of failures that take down the entire blade chassis or require a parts swap such as replacing a backplane/midplane. I'm specifically looking at the HP c7000 chassis and hope to overcome any issues by setting the chassis up like so:

• 208v power #1, UPS #1, generator #1

- connect 2 chassis power supplies

• 208v power #2 , UPS #2, generator #2

- connect 2 chassis power supplies

• Multiple interconnect modules uplinked via LACP to multiple aggregation switches in a loop, the interconnect internal X-overs will be "blocking", each interconnect switch LACP will be "forwarding"

• Redundant administrator modules

• All servers will have pNICs connected to both interconnect modules

• Implementing multiple chassis

Anybody have this setup properly and still experience chassis or SPOF issues?

Thanks,

Ben

[dconvery]

Ben -

I worked for an HP Partner and I am certified in C Class. Your config is the right way to go. I have installed hundreds of chassis for them and only saw ONE midplane choke back in the day when they were still new.

Dave

************************

"There is an island of opportunity in the middle of every difficulty. Miss that, though, and you're pretty much doomed."

despair.com