Stroller writes: > Like I say, never used RAID on Macs, but a little bit on Linux now (with > at > least 3 different controllers). > > > > When you initialise the array it will wipe out any data on the disks. The > long wait you mentioned was the drives zeroing out. After the > initialisation is completed, the two drives of a RAID1 array will always > (in normal use) contain identical copies of data (subject to the > completion > of writes on each). > > > > The initialisation will (surely) NOT mirror existing data to an array > during creation, although some RAID controllers / userspace utilities > offer > a "migrate" command. > > > > What you could do, for instance, if you have a working installation of > MacOS on drive A, is to build an array Z from drives B & C. Then dd from A > to Z. I have undertaken some variations of this under Linux; depending > upon > the complexity of what you're undertaking it can be either very easy and > straightforward or time-consuming and somewhat harrowing. > > > > You will likely only be able to use the drives when connected to the > controller - i.e. you cannot remove one drive and read it on a different > computer, unless that machine has the same RAID controller. If one drive > fails you will be prompted to replace the drive - you should be able to > work with only the single drive in the meantime. After you replace the > failed drive the array will be rebuilt (copied, effectively) onto the new > disk, and again you should be able to use the system during this rebuild > process (disk operations will be slowed, as your use of the disk is > effectively "shared" with the rebuild). This sounds like what the documentation I have been reading was saying. I guess I will go ahead and see if the disk utility will format it and maybe I will get a usable device out of it. Once that happens, it should work as the raid card, itself, seems to be satisfied. Martin McCormick