We're not sure why that error was happening, but rebooting the webserver fixed it. Not going to try to track it down because we expect to have the new webserver into production before too much longer.
An update on the new database server. It's installed and we inadvertently caused slowdowns and timeouts by testing file transfer speeds between the old and the new. A 1.6GB test file transfered in 40 seconds, and will likely transfer even faster in the evening when the system isn't as busy, meaning it's possible it might only take 10 minutes to copy the whole database over to the new machine. I'm sure the current db's SCSI array will be the bottleneck, and not much of one. The new machine and current one are talking to each other through a separate Gigabit switch.
What was really interesting was that once the test file was on the new machine, it took about 3 seconds to copy it from one partition to another. That is some SICK speed! Looking like they weren't lying when they said each hard drive adds 300Mbps throughput to the array and we've got 6 drives in this array, so 1.8Gbps throughput.
I can't comprehend the amount of traffic it'd take for that kind of hard drive bandwidth to become a bottleneck! Of course, many moons ago, I couldn't imagine how I'd ever use up all of my first 30-meg hard drive.
I've got a smallish database on the new machine now copied from the old and will spend as much of today as possible (and with other hands on my time, that won't be a lot of time) checking out how the new version of SQL Server deals with it, and if it looks like it's doing well, I'll take the production database offline sometime tonight temporarily so I can copy it over for further testing.
