On Wed, 2009-03-11 at 21:12 +0000, Alan Cox wrote: > > If you have hardware RAID, a drive failure shouldn't take the system > > down at all. > > Often not true. It's a lot better with SATA than PATA or SCSI. There are > various ways failed devices can jam up busses and its not unknown for > them to trigger controller bugs even in "brand name" setups. With SCSI > busses it was all too common for a failed drive to jam an entire bus. ---- been there...a Dell server with an adaptec based 3Di controller. Lots of fun. I have had software RAID on SCSI without going down too...I think I was relating my anticipation of likelihood. ---- > Ideally you want the two halves of the mirror on different controllers > but it depends how much resilience you need. As you get more serious it > gets more and more pricy until you end up with the "mirrored servers > separated by at least twice the crater size of a worst case airbus a380 > crash on the data centre" > > and there *are* people who work to that spec ;) ---- I however am not one of them. I have however become fond of serial-attached-scsi Craig -- fedora-list mailing list fedora-list@xxxxxxxxxx To unsubscribe: https://www.redhat.com/mailman/listinfo/fedora-list Guidelines: http://fedoraproject.org/wiki/Communicate/MailingListGuidelines