Donald E Haselwood wrote:
>Would RAID-1 be a good idea to avoid the problems from that "yearly" HD crash?
>
>
With today's bit densities and error rates in IDE/SATA drives, you're
kidding yourself if you're _not_ using at least RAID1. Regardless of
wether you're using Linux or any other operating system.
>A partition on my 160 GB drive that holds my Suse9.2 came up with errors.
>Fortunately, it was a partition that held some old backup files so nothing
>important was lost. I do have my important data files well backed up, but if
>the partition holding the system fails it takes a lot of time to get
>everything setup. I was wondering if RAID-1 with a second HD would be better
>than, say, merely copying the whole partition, or some similar backup scheme.
>
>
RAID1 is _not a replacement for backups_. Let me repeat that again:
_always backup your critical data_.
Now on to hardware reliability: modern IDE/SATA drives have a set of
spare blocks that are used to map in place of bad blocks as they are
found. When these bad blocks occur, data is sometimes lost (the old data
is sometimes not retreivable). This is where RAID1 comes in: the other
drive _does_ have the data, and the RAID subsystem will detect the fault
and rebuild the mirror.
If a drive is "going bad", these rebuilds will become more and more
regular until the drive simple refuses to remap any more bad sectors or
goes away entirely.
So, to reiterate: ALWAYS BACKUP YOUR DATA. And if you're going to use
IDE/SATA drives, I strongly suggest _at least_ RAID1 to work around bad
block remapping.
- Ian C. Blenke <ian@blenke.com> http://ian.blenke.com/
-----------------------------------------------------------------------
This list is provided as an unmoderated internet service by Networked
Knowledge Systems (NKS). Views and opinions expressed in messages
posted are those of the author and do not necessarily reflect the
official policy or position of NKS or any of its employees.
This archive was generated by hypermail 2.1.3 : Fri Aug 01 2014 - 20:39:18 EDT