On Wed, 20 Oct 2004 17:29:16 +0100, John P. Looney <valen at tuatha.org> wrote:
> Think about it. the OS does an fsck because it's not sure what's on the
> disk, and what's not. RAID systems have to be consistent also - and if the
> power goes, do you really know if the same data has been written to the
> same disk ?
>> I'm sure it's possible to do journalling for RAID systems, just like it
> can be done for filesystems; as in "I'm about to write these
> blocks".."I've written those blocks, and they are no on both disks".
Yup, see 'dirty region logging' in Veritas VM. I don't recall offhand
if it actually works where the power's failed, or if it only speeds
resyncs where you've done a controlled detach of one side of a mirror
or whatever.
> I think the more expensive hardware raid controllers can do it. Ones with
> battery backed up RAM for write caching.
For those, it's more like the disks aren't inconsistent in the
lost-power case, because the controller writes the cached data out
when it comes back online.
--
David Murphy
Maintained by the ILUG website team. The aim of Linux.ie is to
support and help commercial and private users of Linux in Ireland. You can
display ILUG news in your own webpages, read backend
information to find out how. Networking services kindly provided by HEAnet, server kindly donated by
Dell. Linux is a trademark of Linus Torvalds,
used with permission. No penguins were harmed in the production or maintenance
of this highly praised website. Looking for the
Indian Linux Users' Group? Try here. If you've read all this and aren't a lawyer: you should be!