Re: saving RAIDframe parity accross a panic

[Thor Lancelot Simon <tls%rek.tjls.com@localhost>]

On Tue, Nov 25, 2008 at 05:01:33PM +0100, Geert Hendrickx wrote:
> On Tue, Nov 25, 2008 at 08:48:51AM -0600, Greg Oster wrote:
> > There are plans to make it not have to check as much parity 
> > after a panic, but right now it has to verify all parity...
> 
> For when is this planned?

You too have a text editor!  (Probably even one with convenient
macros for doing things like indenting source code.)

The parity logging code, if it could be quickly made to work at all
(I keep meaning to find the time to try, but my text editor is...busy),
appears to have most of the functionality required for this already.
Specifically, it chops up the disk into many parity regions and tracks
their state independently of each other.

I have no idea if this code would work in a 2-disk "mirror" setup
like many of us use to boot NetBSD.  I suspect that would require
some support in the bootblocks, just as a 2-disk "RAID5" doesn't work
now.

-- 
Thor Lancelot Simon                                        
tls%rek.tjls.com@localhost
    "Even experienced UNIX users occasionally enter rm *.* at the UNIX
     prompt only to realize too late that they have removed the wrong
     segment of the directory structure." - Microsoft WSS whitepaper