Christian Pernegger wrote:

Anything specific wrong with the Maxtors?

No.  I've used Maxtor for a long time and I'm generally happy with them.

They break now and then, but their online warranty system is great.
I've also been treated kindly by their help desk - talked to a cute
gal from Maxtor in Ireland over the phone just yesterday ;-).

Then again, they've just been acquired by Seagate, or so, so things
may change for the worse, who knows.

I'd watch out regarding the Western Digital disks, apparently they
have a bad habit of turning themselves off when used in RAID mode, for
some reason:
http://thread.gmane.org/gmane.linux.kernel.device-mapper.devel/1980/

I read somewhere that this could reduce rebuild time when a "disk"
(partition in this case) is kicked offline because of a timeout or
somesuch. Sounds a bit fishy, which is why I'm asking.

The bitmap code in MD is for fast rebuilding, if you need that.

The point of the whole exercise is that I don't want to be cut off
from my data, just because a part of the host (not the disks) died.

I don't see any large risks of that happening with the setup you've outlined.

And iSCSI sounds way too expensive. :)

I think a host adapter is around $100.