Wojciech Puchar writes:
> > >
> > > raidctl -a /dev/sd0e raid0
> > > raidctl -F component1 raid0
> > >
> > > and it started copying to second disk. after everything done i rebooted O
> S
> > > and... it started to recompute parity once again.
> >
> > Which version of NetBSD?
> 
> 1.5.1
> 
> > > after it i rebooted again (cleanly!!) and again it recomputes parity afte
> r
> > > start.
> > >
> > > what's wrong?!
> >
> > I need to know what version of NetBSD before I can speculate...  do you hav
> e
> 
> / on raid0a and swap on raid0b

So the problem is that swap doesn't get unconfigured, and thus raid0 never 
gets shutdown cleanly.  Some options are:
 1) move the swap to it's own RAID set.  It still won't get shutdown properly, 
but will take less time to rebuild, and you won't have critical filesystems 
on a non-redundant RAID set after a shutdown.

 2) Put a line like:

# KEYWORD: shutdown

in /etc/rc.d/swap1 .  That way when you do a shutdown, the system will 
(hopefully) be able to unconfigure the swap, which should allow the RAID set 
to unconfigure cleanly.  (Hmm.. there might be something else needed here, 
but it's been a while since I last thought about this problem..)

Later...

Greg Oster