gvinum startup problems

[Lukas Ertl]

On Sun, 9 Oct 2005, Daniel Eriksson wrote:

> After the gvinum MFCs earlier today there has been some regression in
> the RELENG_6 branch.

Yes.  Due to the fact that the GEOM framework handles class load events 
differently on a cold boot and after boot, and how it lets these classes 
taste other providers, I had to make a decision.  The only correct way to 
start geom_vinum now is from /boot/loader.conf.

> If gvinum really is this broken, shouldn't it be removed from RELENG_6?

If it is really this broken, you should send patches.

> Or is this just something local on my setup?

That's what I suspect, please see my earlier mail about slices/partitions. 
You may have forgotten to add some offsets to any slices/bsdlabels you 
have configured as vinum drives.

> Oh, and there is still no manpage for gvinum.

I have a man page and other stuff from the Google Summer of Code project, 
which I will integrate after the release.

thanks,
le

-- 
Lukas Ertl                         http://homepage.univie.ac.at/l.ertl/
le at FreeBSD.org                     http://people.freebsd.org/~le/