RAID failure update
:: Saturday, May 27th, 2006 @ 12:46:55 am
:: Tags: Computers
After my RAID went critical yesterday morning, I assigned the failed drive as a global spare (for kicks), and the RAID started rebuilding automatically. 30 hours later the rebuild finished, but the machine hung upon running CHKDSK. So I swapped the faulty drive out with a new one, which had arrived in the mail earlier in the day. And so another 30-hr rebuild has been kicked off. The first CHKDSK recovered more than 50mb of corrupted files and directories, and I’m afraid of what running another instance will find, once the current rebuilt is done.
Between my last backup and the files that exist on my other machine(s), I don’t think I have lost any data. But we’ll see…
This is lame: Maxtor requires that you run a diagnostic tool in order to get an online RMA for defective drives. The problem is, the stupid tool builds a bootable floppy drive! Who uses floppy drives these days!? Floppy-drive-less, I’m going to have to call support for my RMA on Tuesday, since Monday is a national holiday.
Repeat after me: I will never buy Maxtor or Western Digital drives again. I will never buy Maxtor or Western Digital drives again. I will never buy Maxtor or Western Digital drives again.
I’ve already stopped buying WD drives. But I had to replace the failed Maxtor with another MAxtor, and with Seagate’s acquisition of Maxtor, we may have to cross Seagate off of the list as well (even though they’re at the top of my list right now, along with Hitachi).
stick with Hitachi. Our lab testing shows hitachi drives to be out on top. for SCSI or FC drives, Seagate edges out Hitachi by a tad. If we could only have cheap, fast flash drives. next gen hybrid flash/disk drives. 2007 ;)
I don’t buy WD drives anymore and the two I still have in use I essentally treat them as if they could fail at any moment (ie. backups of all data or data that can be lost are on those drives).
Tend to stick with Seagate, Hitachi or for smooth silent read/writes Samsung (spinpoint series).
Hope there is no data loss.