* Paranoid read mode for raid5/6
@ 2007-06-11 14:56 Mattias Wadenstein
2007-06-11 16:39 ` Dan Williams
0 siblings, 1 reply; 2+ messages in thread
From: Mattias Wadenstein @ 2007-06-11 14:56 UTC (permalink / raw)
To: linux-raid
Hi *,
Is there way to tell md to be paranoid and verify the parity for raid5/6
on every read? I guess this would come with a (significant) performance
hit, but sometimes that's not a big deal (unlike disks scrambling your
data).
Also, regarding data paranoia, for check/repair of a raid6, is the effort
made to figure out which is the misbehaving participant device, or is
parity just blindly recalculated?
/Mattias Wadenstein
^ permalink raw reply [flat|nested] 2+ messages in thread
* Re: Paranoid read mode for raid5/6
2007-06-11 14:56 Paranoid read mode for raid5/6 Mattias Wadenstein
@ 2007-06-11 16:39 ` Dan Williams
0 siblings, 0 replies; 2+ messages in thread
From: Dan Williams @ 2007-06-11 16:39 UTC (permalink / raw)
To: Mattias Wadenstein; +Cc: linux-raid
On 6/11/07, Mattias Wadenstein <maswan@acc.umu.se> wrote:
> Hi *,
>
> Is there way to tell md to be paranoid and verify the parity for raid5/6
> on every read? I guess this would come with a (significant) performance
> hit, but sometimes that's not a big deal (unlike disks scrambling your
> data).
>
> Also, regarding data paranoia, for check/repair of a raid6, is the effort
> made to figure out which is the misbehaving participant device, or is
> parity just blindly recalculated?
>
This is something that is not available today, but is something I am
looking to implement. Especially for the raid6 case where hpa showed
that not only can corruption be detected, it can be corrected.
> /Mattias Wadenstein
Dan
^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2007-06-11 16:39 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2007-06-11 14:56 Paranoid read mode for raid5/6 Mattias Wadenstein
2007-06-11 16:39 ` Dan Williams
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).