Paul B Dokas writes:
> On Tue, 6 Apr 1999, Greg Oster wrote:
> >
> > > RAIDFRAME: protectedSectors is 64
> > > raid0: Component /dev/sd0e being configured at row: 0 col: 0
> > >          Row: 0 Column: 0 Num Rows: 1 Num Columns: 2
> > >          Version: 1 Serial Number: 100023 Mod Counter: 183
> > >          Clean: 0 Status: 0
> > > /dev/sd0e is not clean!
> > 
> > Gack.  Clean bits don't mean a whole lot for RAID0, so perhaps 
> > the error message here is a bit bogus..
> 
> Hmm, that's true.  Perhaps the message just needs to go away for RAID 0.

That's what I'm thinking...

> > [snip]
> > > However, everything still works just fine, even though the system demands
> > > that the RAID is not clean.
> > > 
> > > Anyone know of a way to mark the RAID as clean?  I've RTFM'd and tried
> > > lots of different raidctl commands.  None have managed to set the clean
> > > bit.
> > 
> > Doing a:
> > 
> >   raidctl -i raid0
> > 
> > should make it happier.  The initializing of parity should trivially 
> > succeed (as there is no parity) and after that the kernel shouldn't complai
> n
> > about the RAID set not being clean... 
> > 
> > Please let me know if the above doesn't fix things, as if it doesn't, I'll 
> > need to do some digging...
> 
> caligula# raidctl -i raid0
> Initiating re-write of parity
> raidctl: ioctl (RAIDFRAME_REWRITEPARITY) failed: Invalid argument
> 
> 
> I've done this before.  It makes to difference.  The clean bits have
> not been set.

Ya.  The EINVAL gets returned before it even *tries* to trivially re-write 
parity.  

> As you mentioned above, the clean-bit doesn't make sense in RAID 0 

Arguably it makes as much sense as RAID 0 being called "redundant"  :->

> (at
> least not in the same way as it does in RAID 1 and up).  Perhaps it needs
> to be trivially set and/or ignored in RAID 0.

Yup.  Looks like I have something else to take care of this evening :-)

[I'm glad to see that people are actually using the RAID stuff and 
providing feedback.  I'd be interested in hearing (perhaps off-line)
any other experiences/feedback that people have with respect to
RAIDframe and NetBSD.  Thanks. ] 

Later...

Greg Oster