Let me tell you what doesn't work

This should have been a warning. I am stupid for not paying more attention to what that guy was saying.

Let me start at the beginning. We needed more storage- 4TB more storage. We have been ordered by the regents to keep a backup of all "official records." Since no one knows exactly what an "official record" is, but we keep being told it includes all research- or teaching-related materials, we decided we had to back up all of our users' desktop data. So we started deploying the DLO agent on our systems, targeting the nearly 1TB U320 direct attached RAID 5-based SMB volumes for storage space. With over 300 desktops in the college, that quickly filled up. So we started looking at SAN solutions. It's tough when you are limited to a ~$7,000 budget. Fiber channel was out, for obvious cost-related reasons. EMC has some iSCSI products, but they are expensive to upgrade. EqualLogic has some terrific stuff, but they cost as much as a Porsche, and you need two of them to take advantage of the cool replication feature. I just wanted a big fat disk on a gigabit VLAN. We finally decided on this puppy: the Promise Vtrak M500i. 15 SATA II slots, two gigabit iSCSI ports and a management port for ~$4,000. We put 8 Seagate 7200.9 500GB SATA II drives in it, for about $7,000 total. Plugged it in to the gigabit switch, connected it to the spare NICs on a couple of the servers, and provisioned space for a couple RAID 5 arrays. Formatted the logical disks and moved the DLO storage over there. All was well with the world for about 24 hours, then the thing started loop rebooting. You couldn't get at the web interface, all you could get to was the serial console. I finally managed to ctrl-C it into giving me a boot loader prompt, flashed the firmware, cleared out the nvram (which lost all my settings! Agggh!) and rebooted it. Keep in mind none of how to do this is in the horrendously poor documentation for the enclosure. I had to figure it all out by the seat of my pants. This process fixed the problem for another 24 hours. Then the same thing started happening again. This time, when it started loop rebooting, the status light on drive #3 stayed unlit. I suspected this drive was bad, took it out, and the array failed over to the global hot spare. Everything has been fine since.

I was very unhappy, contacted Promise and they told me that the drive had firmware that was "too new." What the heck? The drive is on their compatibility document, but they list firmware 3.AAB. These drives have 3.AAD. That should not make a bit of difference. I suspect the drive is physically bad. Still, that should not make the entire enclosure freak out. What a bunch of hooey.