A SAN is a fragile thing, really.

So in a test lab life is not like in an enterprise environment. I cannot be, of course. Our job is to pull a bunch of pieces of hardware and software together to make a test configuration, run our tests, and then rip it all apart and start over.

It occurs to me after this most recent test that SANs just weren't made for you to swap 2-3 core infrastructure pieces in and out of each day.

My most recent testing - for "Storage Security" has involved me pulling pieces - including switches - in and out of our test SAN. It has been down hard several times during the course of the review.

We get through it, of course. We've had power outages, failed switch configs, interoperability issues, and even a broken CD ROM drive on a host, but for each one we find a fix and move on.

The vendors have all been pretty understanding - most understanding goes to Protegrity, who sat waiting for me to get the SAN working correctly again and the databases reinstalled until 6 PM, then worked with me late into the night to start the tests, only to come back at 9 am and finish up testing.

But it makes me glad that changes to storage infrastructure were infrequent in my enterprise experience. I wouldn't recommend making unnecessary wholesale changes to yours... Not that most people do.

Meanwhile, the EqualLogic iSCSI SAN just chugs along. It's got some issues, but there are no SAN switches to fail or have interoperability problems, there's actually not much that can go wrong with it. Hopefully now that I've said that, when I head back to the lab this afternoon it won't have burned out both power supplies....