[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Bug#857916: linux-image-4.9.0-2-amd64: btrfs modifying wrong mount point data upon new mount



Control: tag -1 upstream

On Thu, 2017-03-16 at 11:49 +0100, Philipp Marek wrote:
> Package: src:linux
> Version: 4.9.13-1
> Severity: normal
> 
> I'm using DRBD to keep a backup of my root filesystem (multiple snapshots 
> on remote machine).
> 
> Now I wanted to access an old version of a file; so I got me access to the 
> right snapshot (again via DRBD), and tried to mount the filesystem.
> 
> First I noticed that the actually passed in block device wasn't even used; 
> btrfs noticed that a filesystem with the same UUID is already mounted, and 
> so "helpfully" just mounted the (*wrong*) block device at the new location.
[...]

btrfs includes its own logical volume management and can spread a
filesystem across multiple devices, so in general it has to use UUIDs
to find the devices for a filesytem.

I agree that this interacts poorly with block-level snapshot
mechanisms.  If you want to have multiple versions of a btrfs
filesystem, I think you have to use btrfs snapshots.

Ben.

-- 
Ben Hutchings
The two most common things in the universe are hydrogen and stupidity.

Attachment: signature.asc
Description: This is a digitally signed message part


Reply to: