Subject: Re: RAIDframe works fine, but I'm wondering...
To: Greg Oster <oster@cs.usask.ca>
From: Anders Dinsen <anders2@dinsen.net>
List: netbsd-users
Date: 07/24/2000 20:54:18
On Sun, 23 Jul 2000, Greg Oster wrote:
> If you're wanting to do pure concatentation of different-sized disks, 
> then you'll have to use ccd, since RAIDframe doesn't handle that... 
> If the disks are all the same, then using a RAID 0 w/ RAIDframe should work 
> just fine.. 

No, they're identical and I want to increase performance, so RAID 0 should
work.

> > I've experimented a using two vnode-disks, with success (except the kernel
> > panic'ed when they were only 1MB large - works fine with two 20 MB disks?
> > anyone know if this is a known problem or should I submit a pr?)
> 
> Somewhat known :)  But it wouldn't hurt to file a PR anyway... (Is the panic 
> RAIDframe-related, or something else? )

I have lost the exact message, because it was only on the console, but it
did mention aha1542.c and could'nt sync the disks because SCSI requests
timed out. So either, the driver went into an unknown state or the adapter
did. 

It could therefore not write a core image either.

I will try to reproduce it.

> Remember to do both 'raidctl -I' and 'raidctl -i' too... then, copy rfconfig 
> to /etc/raid0.conf , and /etc/rc should use /etc/raid0.conf to automatically 
> configure the RAID set at boot... (I *think* the right stuff to do this
> is in /etc/rc for 1.4.1, but I don't remember any more... :-/  If it isn't
> you can certainly borrow those parts from 1.4.2.)

Well, my /etc is mostly leftovers from my 1.3 installation, but I have
taken the raid[0-3].conf detection and configuration from the 1.4.2
etc.tgz.

> Later...

See you on the other side of this adventure :-) (my /home and most other
user files are on the new disk).

- Anders