[ On Friday, April 19, 2002 at 11:16:47 (-0600), Greg Oster wrote: ]
> Subject: Re: RAIDframe questions 
>
> Jeff Rizzo writes:
> > OK, I'm doing some of that tweaking now, playing with different 
> > block/frag sizes and stripe widths.  One issue that's come up is
> > that I'm not entirely sure what the accepted way of recreating a 
> > raid set with different params is... when I change something in the
> > conf file (like stripe width), and reconfigure the raid set, it
> > claims the parity's still good, which seems counterintuitive to
> > me.  Perhaps I need to erase the component labels? 
> 
> Yup. 
> 
> > Is there
> > any way to do this short of dd'ing the whole disk with zeros?
> 
> You just need to nuke the first 32 blocks (16K) at an offset of 16K from the 
> beginning of the partition.  (if the partition doesn't start at block 0, then 
> you can just nuke the first 32K without worrying about disklabels and such).

One of the major problems I had with VINUM on FreeBSD was re-doing
broken configurations.  It took me a long time to realise it too was
pulling "garbage" off the disk from magic places.  It took even longer
to work this out than you might expect too -- the system kept crashing
as there seemed to be little validation of the configuration information
that was taken off the disk, and indeed little validation of what was
put there in the first place.  A screwed up config would hose the system
completely.  It works very well once you forcibly zap the bad/broken
component labels though....

I found that RAIDframe was a lot better at handling both user errors and
random garbage on the disk.  If I remember correctly it would refuse to
do obviously stupid/impossible things whether told to do so from the
user-land config or from the component labels.  Although I didn't play
around a lot with it (I wanted to get the system up and usable! ;-), I
was under the impression that 'raidctl -C' would ignore any existing
configuration in the component labels, and if followed by a 'raidctl -i'
would always re-write the parity data.

Is this not true?  Surely there's a better way to force reconfiguration
of the component labels than to dd /dev/zero to them....

-- 
								Greg A. Woods

+1 416 218-0098;  <gwoods@acm.org>;  <g.a.woods@ieee.org>;  <woods@robohack.ca>
Planix, Inc. <woods@planix.com>; VE3TCP; Secrets of the Weird <woods@weird.com>