Is this a gmirror bug?

[Chris Cowart]

Wojciech Puchar wrote:
> > I've seen this kind of thing appear in my df output:
> > linprocfs                   4       4         0   100%    /proc
> >
> > /dev/mirror/gm0d      4058062 -377792   4111210   -10%    /tmp
> >
> > /dev/mirror/gm0e     15231278 -113942  14126718    -1%    /var
> >
> > /dev/ad10s3e        121487580       4 111768570     0%    /v3
> >
> > /dev/ad8s3e         121487580       4 111768570     0%    /v2
> >
> > /dev/ad6s3e         121487580       4 111768570     0%    /v1
> >
> > /dev/ad4s3e         121487580       4 111768570     0%    /v0
> >
> >
> >
> > It's showing that two partitions in my gm0 partition are below 0%
> > capacity. This is clearly wrong, but what does it mean?
> >
> it has nothing to do with gmirror - no matter if it's virtual disk (gm0 
> that case) or physical, partition or not, it's just block device to UFS.
> 
> definitely it is some problem but with UFS here.
> 
> unmount this filesystems and do fsck_ffs -y on them

Nothing is wrong.

10% of the disk space is reserved for the superuser. The 10% free
mark is what shows as 0% in df. If you're negative, it means you've
tapped into the super-user reserve. This is not good, because it means
you've lost a lot of the FS-level optimizations from UFS.

-- 
Chris Cowart
Network Technical Lead
Network & Infrastructure Services, RSSP-IT
UC Berkeley
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 834 bytes
Desc: not available
Url : http://lists.freebsd.org/pipermail/freebsd-questions/attachments/20090526/dca3354a/attachment.pgp