RE: oops umounting full LVM snapshots

[Eric Sandeen <sandeen@sgi.com>]

The issue here is changing the quota state on a read-only device - i.e.
if quota's on in the superblock but mounted w/o quota, or vice versa.
We can't do this on a read-only device because this will require a
transaction.

It's not the VFS flags we're worried about, it's the whether the device
itself is read-only (as in the LVM snapshot case).

So if we're trying to change quota state on a read-only device, we
should fail.

I have the basic code going to detect the situation, but I'm having
trouble erroring out of the mount path cleanly, now.  Each solution
brings new problems. :)

The quick fix for those of you experiencing this is to mount your
snapshots with the same quota options as the original filesystem.

-Eric


On Thu, 2002-02-28 at 22:18, FORRESTER,JUSTIN (HP-Loveland,ex1) wrote:
> 
> I was able to get rid of the problem with a simple hack to clear out the
> quota flags in the [in-core] sb if we're mounted ro.  The fs shutdown
> problem went away, as did the subsequent oops.  fyi.
> 
> Justin

-- 
Eric Sandeen      XFS for Linux     http://oss.sgi.com/projects/xfs
sandeen@sgi.com   SGI, Inc.