Subject: Re: raidframe on a pmax?
To: Simon Burge <simonb@NetBSD.ORG>
From: Greg Oster <oster@cs.usask.ca>
List: current-users
Date: 08/03/1999 11:43:09
Simon Burge writes:
> Greg Oster wrote:
> > That's about right... You'll notice a performance improvement for writes 
> > with a -current kernel..
> 
> I can live with the write-performance for now...  Is the improvement in
> -current a percentage, an order of magnitude, something else?

Depends on the box... Bonnie's per-char output seems to just about double, and 
the per-block output sees something like a 10 to 100% improvement too (depends 
on the disks, the types of disks, etc... but it's noticable).

> > > My primary concern at this stage is the time it takes to rewrite the
> > > parity.  Has any thought been given to backgrounding this process so
> > > that the raid set operates in degraded mode until parity is recomputed?
> > > (No I haven't read the RAIDframe papers yet :).
> > 
> > :-) Just run 'raidctl -i raid0' in the background.  You can still fsck, 
> > mount, use, etc. the filesystem while the parity is begin computed -- it's 
> > just that it's somewhat riskier, of course.  You actually don't want the ra
> id 
> > set to be operating in degraded mode before parity is computed, as degraded
>  
> > mode implies that you may be using the parity bits to compute data bits -- 
> but 
> > at that point you don't know that the parity bits are good! (if you knew th
> e 
> > parity bits were good, you wouldn't be looking to recompute them :) ).
> > While the parity is begin recomputed, the real data blocks will be being re
> ad 
> > on reads, and on writes, the real data blocks will be written, and the 
> > associated parity blocks will be updated as well.  The parity re-write can 
> > operate independent of all of this, and can do it's job even if you're work
> ing 
> > with the RAID set.
> 
> This isn't mentioned in the man page for raidctl.  In the description of
> the -i option, it may pay to mention that this task can be backgrounded
> (with the appropriate warnings). 

Good point.

> /etc/rc support for autodetecting
> non-clean parity and starting a "raidctl -i" in the background would be
> way-cool - I'd play with this myself, but the only box I've got with
> RAIDframe is a production box, so this isn't an option :-(

No worries.. cleaning up /etc/rc support as you've suggested is on the todo 
list. 

Later...

Greg Oster