Re: md(4) broken on CURRENT?

On 06/06/07, youshi10_at_u.washington.edu <youshi10_at_u.washington.edu> wrote:

> > On Thu, 7 Jun 2007, Kazuaki ODA wrote:
> >
> >> On CURRENT amd64 cvsup'ed yesterday:
> >>
> >> # dd if=/dev/zero of=/var/tmp/disk1.img bs=1k count=1m
> >> 1048576+0 records in
> >> 1048576+0 records out
> >> 1073741824 bytes transferred in 29.206387 secs (36763939 bytes/sec)
> >> # dd if=/dev/zero of=/var/tmp/disk2.img bs=1k count=1m
> >> 1048576+0 records in
> >> 1048576+0 records out
> >> 1073741824 bytes transferred in 27.405935 secs (39179171 bytes/sec)
> >> # mdconfig -a -t vnode -f /var/tmp/disk1.img
> >> md0
> >> # mdconfig -a -t vnode -f /var/tmp/disk2.img
> >> md1
> >> # mdconfig -l
> >> <Nothing outputted>
> >> # gmirror load
> >> # gmirror label data md0 md1
> >> # gmirror status
> >> gmirror: Cannot get GEOM tree: Unknown error: -1
> >>
> >> Hmm, is md(4) broken on CURRENT?  Or is this a problem only on amd64?
> >>
. . .
> Works for me on 7-CURRENT/amd64 cvsup'ed and completely rebuilt Sunday.
> -Garrett

With csup as of late in the day 07jun (built last night),
on i386, it works as expected.

-- 
--